About the Canadian Wildlife Service

Transcription

1

2 About the Canadian Wildlife Service Environment Canada s Canadian Wildlife Service is responsible for wildlife matters that are related to species of federal concern. This includes protection and management of migratory birds as well as nationally significant wildlife habitat, endangered species, control of international trade in endangered species and applied science on wildlife issues of national importance.

3 Executive Summary An important challenge for Canadian society in addressing the natural environment, one of the three elements of sustainability along with society and economy, is to ensure that there is adequate wetland, riparian, forest and grassland habitat to sustain minimum viable wildlife populations and help maintain selected ecosystem functions and attributes. The goal of the How Much Habitat is Enough? series is to provide science-based information and guidelines related to these natural systems and the biodiversity for which we all are the stewards. The information is tailored to be of use to government and non-government restoration practitioners, planners and others involved in natural heritage planning, conservation and preservation. The guidelines concentrated on providing information to assist decision makers in Great Lakes Areas of Concern in the setting and achievement of delisting criteria concerning fish and wildlife habitat beneficial-use impairments, and for post-delisting to provide further guidance on habitat restoration. An assessment of the guidelines (First Edition) in 2002 demonstrated that they were well-used within, but also outside of, Areas of Concern. The assessment indicated that they were indeed used as originally envisioned as a guide to set restoration targets and restoration project locations, and also as a sciencebased reference for natural heritage practitioners. A Second Edition was prepared and released in Regular updates are necessary to ensure that current research and scientific literature is incorporated and to provide an opportunity to expose new ways of considering old problems. This has also provided Environment Canada with an opportunity to invite contributions from a wider selection of topic experts. In this, the Third Edition, there are 21 wetland, riparian, forest and grassland habitat guidelines and accompanying rationales. These are typically supported with a discussion addressing additional potential guidelines or issues. The grassland section is a new chapter that represents a first attempt at addressing these complex habitats. More than any other habitat type, grasslands are very dependent on human activity. This section is sure to evolve in the future as many of the newly listed species of conservation concern are found within grassland habitats. In the forest section, there has been an important shift toward a risk-based approach, one that is likely to be applied in other habitats in future years. Overall, the guidelines have expanded significantly both in terms of the supporting discussions within each topic and the base of scientific literature that has been reviewed. The framework of guidelines and supporting text that comprise How Much Habitat is Enough? are effectively an open file meant to be built upon and to be adapted according to historical and current local conditions. This framework will hopefully continue to serve as a starting point to develop strategies to conserve habitat, develop natural heritage systems and discuss research needs around those habitats. This is in the spirit of keeping common species common, while restoring ecosystems for those species most at risk the sensitive, the endangered, the threatened and the rare. How Much Habitat is Enough? 1

4 Acknowledgements for the Third Edition How Much Habitat is Enough? (A Framework for Guiding Habitat Rehabilitation in Great Lakes Areas of Concern) (1998) was initially guided and championed by the Ontario Ministry of the Environment, the Canadian Wildlife Service of Environment Canada and the Ontario Ministry of Natural Resources. Subsequently, it has evolved into a living document in response to challenges, questions, experience and information brought forward by an active cross-section of individuals, agencies and organizations. These numerous contributions are greatly appreciated and acknowledged with this third edition. This new edition further adds to the extensive list of individuals who have worked on this publication since its beginning in 1995 and on the first and second editions (1998, 2004). There have been many authors, contributors and reviewers over the past 18 years drawing from private consultants, federal and provincial agencies, Conservation Authorities, academia, and conservation organizations. Funding support has come from the Great Lakes Sustainability Fund of Environment Canada, the Ontario Ministry of the Environment and the Canadian Wildlife Service of Environment Canada as well as much support in the form of many in-kind reviews and advice. For the Third Edition: Editors: Graham Bryan and Brian Henshaw Nancy Patterson provides guidance, direction and editorial oversight for the How Much Habitat is Enough? series. Lesley Davy has provided coordination, additional review, and research and editing of the third edition. Reviewers and additional contributors include: Aquila Conservation & Environment Consulting David Kirk; Beacon Environmental Margot Ursic; Bird Studies Canada Debbie Badzinski, Audrey Heagy, Jon McCracken; Canadian Wildlife Service John Brett, Mike Cadman, Lesley Dunn, Christian Friis, Kevin Hannah, Nancy Patterson, Daniel Rokitnicki-Wojcik; Central Lake Ontario Conservation Authority Heather Brooks, Kathy Luttrell, Jackie Scott; Conservation Ontario Bonnie Fox, Don Pearson, Leslie Rich, Jo-Anne Rzadki; Couchiching Conservancy Ron Reid; Credit Valley Conservation Dan Banks, Paul Biscaia, Kirk Bowers, Mark Eastman, Kate Hayes, Rod Krick, Liam Marray, Bob Morris, Judi Orendorff, Aviva Patel, Kamal Paudel, Yvette Roy, Saleh Sebti, Scott Sampson, Paul Tripodo, Christine Zimmer; Environment Canada Anie Cyr, Kathy Lindsay; Ganaraska Region Conservation Authority Ken Towle; Gray Owl Environmental Inc. Al Sandilands; Kilgour and Associates Anthony Francis; Ontario Ministry of Agriculture, Food and Rural Affairs Paul Smith; Ontario Ministry of Natural Resources Mike McMurtry, Angus Norman, Brian Potter, Danijela Puric-Mladenovic, Les Stanfield, Allen Woodliffe; Nature Conservancy of Canada Dan Kraus; Ontario Power Generation Steve Hounsell; Paul Keddy; Toronto Region Conservation Dan Clayton, Noah Gaetz, Dave Lawrie, Ryan Ness, Lionel Normand, Paul Prior, Christine Tu, Ron Wu-Winter; Trent University Tom Whillans; and Upper Thames River Conservation Authority Cathy Quinlan. 2 How Much Habitat is Enough?

5 Table of Contents 1. How Much Habitat is Enough? Third Edition Introduction Habitat Guidelines Wetland Habitat Guidelines Percent Wetlands in the Watershed and Subwatersheds Wetland Location Amount of Natural Vegetation Adjacent to the Wetland Wetland Area, Shape and Diversity Wetland Proximity Wetland Restoration Riparian and Watershed Habitat Guidelines Width of Natural Vegetation Adjacent to Stream Percent of Stream Length Naturally Vegetated Percent of an Urbanizing Watershed that Is Impervious Additional Riparian and Watershed Considerations Fish Community Targets Forest Habitat Guidelines Percent Forest Cover Area of Largest Forest Patch Percent of Watershed that Is Forest Cover 100 Metres from Edge Forest Shape Proximity to Other Forested Patches Fragmented Landscapes and the Role of Corridors Forest Quality: Species Composition and Age Structure Grassland Habitat Guidelines Where to Protect and Restore Habitat Type and Area Landscape Configuration, Heterogeneity and Connectivity Patch Size Landscape Heterogeneity Additional Grassland Considerations Cited References How Much Habitat is Enough? 3

6 1. How Much Habitat is Enough? Third Edition Introduction How Much Habitat is Enough? provides sciencebased guidance to conserve and restore habitat for migratory birds, species at risk and other wildlife species within the settled landscapes of the lower Great Lakes and Mixedwood Plains. This biophysical area crosses provincial and international borders: in southern Ontario it forms an area south and east of the Canadian Shield. The landscape of much of the world has irreversibly changed with increasing competition for land and resources. In many areas, the sustainable management of landscapes that considers the natural environment, society and economy is now commonly accepted as the goal in both land use planning and policy (North-South Environmental 2010). How Much Habitat is Enough? plays a role in that discussion by providing guidance for the restoration and conservation of wildlife habitat: in particular habitat of wildlife species under Environment Canada s mandate. In order to contribute best, general guidance is provided at the scale of habitat patches, watersheds and subwatersheds, reflecting the scale at which land use planning occurs in the settled landscapes of southern Ontario. The guidelines in this report are a nonexclusive contribution as to how that landscape could look, reflecting one approach at a particular scale. complexity of life and the multiple and often unknown linkages that allow species to flourish. The Third Edition of How Much Habitat is Enough? contains new and updated science that has been used to provide new guidance or revise existing guidelines. This edition builds on previous versions, combining and removing some guidelines to reflect better the current body of knowledge. And where there is now a sufficient amount of supporting science, requests for guidance on topics such as grassland habitat have been addressed. How to use this guide: Principles and considerations How Much Habitat is Enough? was developed through the 1990s and published in 1998 as A Framework for Guiding Habitat Rehabilitation in Great Lakes Areas of Concern. This framework, comprised of guidelines and supporting information, has been used extensively for land use, habitat restoration and land securement planning as well as for policy development, as a post-secondary classroom resource and as a general conservation primer. The basic principles the framework was built upon have endured and lessons have been learned over the years. It is essential to consider these principles and lessons before applying the guidelines. The guidelines address the size and configuration of habitat patches and the overall quantity of habitat across a landscape for multiple species. It is an inclusive approach that allows for increased conservation of ecosystems. Such a systematic approach better captures the 4 How Much Habitat is Enough?

7 The basics There are four habitat categories addressed by the guidelines: wetlands, riparian and watershed, forests, and grasslands. In reality these habitats overlap and are separated only to provide clear guiding principles. For example, a swamp could simultaneously provide forest- and wetlandrelated ecological functions and can be considered as forest and as wetland under these guidelines. Also, the individual guidelines and principles should not be applied separately but as a suite of interacting and interdependent guidance. For example, guidance on grassland patch size is best addressed in context of the overarching suggestion to target existing and potential grassland landscapes. Where habitats are not defined, please refer to the community class descriptions found within Lee et al. (1998). Science The guidelines are based on scientific literature with an emphasis on published and unpublished studies from eastern North America, south of the Boreal Shield, and in particular the lower Great Lakes region. Supporting information and references are provided with each guideline, and the reader is urged to refer to the original studies and to adapt, and not necessarily adopt, the guidelines. Before applying the guidelines, consider that the advice provided emphasizes the habitat needs Figure 1. A conservation science primer from the Canadian Wildlife Service of migratory birds and other species of federal concern, such as species at risk, and that the guidelines also reflect the limits of existing scientific knowledge. To help bridge knowledge gaps regarding species of federal concern, science on species that are not specifically under the federal mandate has also been used to help formulate the guidelines. Regardless, restoring and conserving habitat using the guidelines will contribute to the maintenance of multiple ecological functions and health across various mandates, along with associated derived goods and services for humans. This is discussed further in the text box Why birds? Additional general guidance regarding conservation biology and landscape ecology is provided by sources such as Environment Canada s Beyond Islands of Green report ( default.asp?lang=en& xml=1 CEC82DE-8E6C- 4A51-A0A5-48A2A6FD078D), and information on the needs of migratory birds is provided in Bird Conservation Region (BCR) Strategies ( lang=en&n=1d15657a-1). How Much Habitat is Enough? 5

8 Guidelines are not binding targets The framework is not legislative and should be viewed as a means to guide, not dictate, local decisions, by providing planners, rehabilitation teams and other decision makers with the best available information. Direction regarding habitat for specific species at specific locations is provided through various provincial and federal policies and statutes including the federal Species at Risk Act and Migratory Birds Convention Act, 1994 and Ontario s Endangered Species Act, 2007, Fish and Wildlife Conservation Act and Planning Act. The general guidance provided within How Much Habitat is Enough? is secondary to direction provided under such statutes. Conserve it first The conservation of existing habitat must remain the most important ecological planning activity in any jurisdiction. Restoration will always be necessary to have a fully functioning natural heritage system in degraded landscapes, but protecting existing habitat is far more efficient and more effective. The guidelines are minimums The How Much Habitat is Enough? guidelines are intended as minimum ecological requirements with the objective to maintain wildlife populations at levels that would prevent local extirpations of species. However, the preference is to maintain population levels to provide better for long-term species persistence: this preference is reflected in the Third Edition in the percent forest cover guideline, which has been rewritten to address risk associated with different forest cover amounts. Generally, a greater diversity and amount of habitat than the minimum will almost always be more beneficial in terms of supporting healthy species populations and a wider range of ecological functions. It should be noted that How Much Habitat is Enough? should not be used to set lower habitat targets for landscapes that Why birds? An example of the limits of scientific knowledge and the value of focal species Forest birds are often used as indicators of the quality of the landscape because they are more easily surveyed, and more is known about their habitat requirements and distribution than other wildlife groups. Likewise, the protection and regulation of migratory birds is a major part of the Canadian Wildlife Service s mandate. The guidelines in How Much Habitat is Enough? reflect the general minimum habitat requirements of the different bird pillars or guilds, encompassing priority species within the Ontario portion of BCR 13. Where there are specialized habitat requirements for individual species, these needs are noted in the Bird Conservation Strategy for Ontario s Bird Conservation Region 13: Lower Great Lakes/St. Lawrence Plain (Environment Canada 2012) or within relevant recovery strategies or action plans (see the SARA registry: Less is known about the sensitivity of invertebrates, amphibians, reptiles, plants and small mammals to, for example, forest fragmentation, although amphibians and reptiles have been receiving more research attention over the past decade or so. In the absence of this knowledge, forest birds serve as a useful indicator of forest and ecosystem health. They can be seen as umbrella species, a term debated within the field of conservation science that refers to species with requirements that overlap requirements for many other species. 6 How Much Habitat is Enough?

9 currently have habitat in excess of the guideline minimums, as often irreversible ecological damage will occur. Removing or degrading habitat, for example moving from 50 to 30% forest cover, will result in reduced wildlife populations, reduced capacity to provide ecological goods and services, and generally lower ecological integrity. The guidelines should be a starting point: where local decisionmaking can provide more habitat than the minimum, then a greater robustness in natural heritage systems can be anticipated. Coalition s Big Picture illustrate the need to look at larger ecosystems and landscapes. Consider landscape context If a planning unit spans across different ecoregions, then consider representing the full range of natural communities that occur in those ecoregions. For example, concentrating habitat conservation and restoration in only one part of a large watershed may result in an underrepresentation of communities and species in other ecoregions within that watershed. Adapt first, adopt second The guidelines are general and not landscape- or watershed-specific. They should be viewed as ecological principles that detailed local scientific and other knowledge, including traditional knowledge, will build upon and inform where appropriate. Local ecological differences such as the location of historic grasslands or original extent of wetlands, and whether precedence will be given to forest, wetland or grassland habitats, need to be considered where appropriate. Look beyond local boundaries Habitat within a planning unit, such as a watershed or municipality, is often ecologically connected with features beyond that unit. In order to promote linkages of habitat between watersheds and across landscapes, surviving habitat corridors and geographic features should be carefully considered. Likewise, local efforts should consider larger-scale planning efforts, such as conservation blueprints and conservation action plans, unique biogeographic features such as the Oak Ridges Moraine, and the role the local landscape plays in the overall diversity and integrity of larger ecoregions and ecozones, such as the Mixedwood Plains. Projects such as the Carolinian Canada Figure 2. Multiple ecoregions in one watershed jurisdiction How Much Habitat is Enough? 7

10 Acknowledge stressors beyond habitat There are additional stressors that affect fish and wildlife populations beyond the loss of habitat, including toxins, nutrient enrichment, disease and invasive species. In several cases, species are at risk not because of direct habitat loss but because of other stressors such as invasive pathogens or species (e.g., Butternut, American Chestnut and ash trees). The importance of the matrix The matrix in landscape ecology typically refers to land cover or land covers that dominate a landscape. In the Mixedwood Plains, the rural areas outside of large cities and towns are predominantly agricultural, dominated by open farmland and interspersed with natural features and small towns and villages. The farmland may be active, lying fallow or regenerating to forest, wetland or other natural land covers. In large cities and towns, a built-up urban matrix dominates. In the Mixedwood Plains and other settled landscapes, protected natural areas, natural heritage features (such as forest patches) or natural heritage systems (interconnected or linked natural features) are embedded within these urban and agricultural matrices. Earlier scientific literature largely disregarded the influence of the matrix on species within natural areas in fragmented landscapes. However, there is a mounting body of evidence suggesting the nature of the matrix can have a profound effect on habitat use by different species, particularly in highly fragmented landscapes. It can do this by (a) directly influencing the dynamics within the natural habitat patch itself, and (b) influencing the ability of species to move between patches (Ewers and Didham 2006). Studies to date in fragmented landscapes have shown that effects are generally greater in urban than in rural areas (Borgmann and Rodewald 2004; Dunford and Freemark 2004; Hansen et al. 2005), and that successional habitats in rural landscapes can provide important supporting habitats for forest species (Gibbs et al. 2005; Milne and Bennett 2007). This means that in much of the Mixedwoods Plains Ecozone, the attributes of the matrix can be as important in influencing species composition and abundance as the attributes of the natural habitat patches themselves. 8 How Much Habitat is Enough?

11 The Third Edition: Emerging considerations Some of the following emerging concepts were noted during the preparation of the Third Edition. Acknowledge the limits of land use planning, restoration and protection The guidance in this report is aimed at a common framework of conservation measures that include land use planning, habitat restoration and protecting natural areas. However, there are forces, such as commodity markets, that influence the landscape and do not easily fit within this conservation framework. For example, agricultural practices create habitat for grassland birds in the form of hay or pasture. These practices are driven by factors such as market forces and consumer demand. Land use planning can allow uses such as planting hay and grazing, but generally it cannot force land to be used in a particular way. Species at risk The guidelines are intended to help keep common species common, and the general guidance provided should not be seen as a substitute for identifying specific species-at- risk habitat under the Species at Risk Act (Government of Canada 2002) or regulated habitat under the Endangered Species Act, 2007 (Government of Ontario 2007). It is intended that the guidance in this report will help to protect and conserve species at risk by restoring ecosystems upon which those species rely. Reference points, populations and change The current settled landscapes of northeastern North America have undergone great change since before European contact. After the retreat of the great ice sheets and prior to European contact, southern Ontario was a forest biome with vast swamps and abundant marshes that supported extensive stream systems and abundant populations of forest and wetland species. There was an unknown extent of grassland habitats, many of which were created by the activities of First Nations. Subsequent to settlement by Europeans and with land clearance in the 19th century, an open country matrix arose. Today, the landscape continues to change along with species populations. How Much Habitat is Enough? looks to the dominant habitats associated with these time periods as multiple reference points for restoration efforts. The goal is to incorporate habitat requirements for a diverse suite of species into the current landscape, while factors such as climate change and invasive species continue to challenge and change ecosystems. The guidelines work on the principle that efforts to conserve and restore the extent and integrity of natural and surrogate habitats will increase ecosystem resilience and better allow species to adapt to evolving conditions. How Much Habitat is Enough? 9

12 A note about scales and planning units The guidelines operate at different scales and are applied to different features to reflect specific ecological functions. The wetland and riparian guidelines tend to address the health of aquatic habitat within a watershed; therefore, watersheds are the unit and scale used within these guidelines. The forest and grassland guidelines tend to address terrestrial species needs. These needs are independent of watershed boundaries and are instead described in units such as patches and landscapes, as those are the units where ecological effects will be realized. In addition, many of the guidelines are often described in terms of a watershed or municipal scale, regardless of ecological function, because watersheds and municipalities are common conservation planning units. An example of using units independently of function can be seen in one of the forest guidelines. The guideline calling for 30 to 50% forest cover applies to an area under 500 to 1000 square kilometres. This happens to be roughly the size of two major land use planning units: most rural Ontario municipalities and smaller fourth-order (quaternary) watersheds. While the guideline refers to terrestrial habitat independent of watershed function or municipal boundaries, a watershed or municipality is a similar-sized unit that is useful for planning purposes Km 2 watershed municipality Figure 3. Similar-sized land units that can be used for planning purposes There is also a hierarchy to some of the guidelines. In terms of the forest guidelines, when regional forest cover is below 30 to 50%, then guidelines addressing patch size and configuration gain importance. When forest cover is above 30 to 50%, then the configuration of the habitat is less important for maintaining species richness. 10 How Much Habitat is Enough?

13 Urban areas How Much Habitat is Enough? assumes a predominantly non-urban context. In some of these watersheds, changes to ecosystems have not been so drastic as to preclude rehabilitation of these systems. However, in many urbanized areas, ecosystems have shifted to an entirely new, and often irreversible, ecological state. It will therefore be impossible to fully implement these guidelines in urban areas. For example, a built-up environment will generally not be suitable for many area-sensitive forest birds. The most effective conservation approach would be to identify and protect habitat above minimum levels before urbanization occurs. However, this does not preclude increasing habitat cover in urban areas. Wetlands, forests, grasslands and riparian zones provide many vital ecological services both for wildlife and humans. Even partial implementation of the guidelines will yield ecological benefits. It mean temperature increase of 2.5 to 3 C as well as an increase in annual levels of precipitation in the order of 10% (Expert Panel on Climate Change for Ontario 2009). Habitats in North America and the species they support are already beginning to show responses to climate change. Documented responses include range shifts in some species, earlier calling for some amphibians, and longer stays on their breeding grounds by some wetland birds (2degreesC 2007; Crick 2004; Niven et al. 2009; Varrin et al. 2009). However, for some of the Mixedwood Plains habitats the net impact upon extent and composition remains unclear. For example, some wetland species that are sensitive to small changes in hydrologic regimes may be affected. It is possible that some wetland species will expand their range or populations due to a generally may be far more appropriate to consider new baselines and targets for habitat in urban areas. There may also be discussions as to the need to compensate elsewhere in a region for habitat loss due to urbanization within that region, which would affect habitat targets set outside urban areas. Climate change Although there are many models and predictions, current data suggest that over the next 40 or so years there will be an average Figure 4. A guide addressing habitat needs in urban areas warmer and wetter climate. The full effects of climate change remain to be determined, and as the climate change models become more refined it will be possible to better predict anticipated effects on a regional scale, and for different habitats. A precautionary approach is to strive toward protecting and restoring more complete ecosystems with greater integrity that will be more resilient to change. This requires venturing beyond the minimum amounts of forests, wetlands, grasslands and riparian areas required to maintain species populations above extinction thresholds. How Much Habitat is Enough? 11

14 What about the Canadian Shield? The environment of the Canadian Shield contrasts greatly with the settled landscapes of the Mixedwood Plains of southern and eastern Ontario. The Shield s complex terrain is characterized by rocky outcrops, relatively thin soils covering Precambrian bedrock, and an abundance of lakes and wetlands. It remains largely forested. In this rugged landscape, with relatively low agricultural potential, the human population density never reached that of areas further to the south. Today, the Shield is in many regards an opposite of the Mixedwood Plains cleared land and human settlements form patches within the forested matrix. In such an environment the approach taken and presented in How Much Habitat is Enough? based as it is on science from the settled agricultural landscapes does not transfer well to the Shield. An area of particular interest to planners and ecologists alike is the southern part of the Shield. South of the French River and Algonquin Park is a distinct transition area bordering the Mixedwood Plains and the Great Lakes. This is an area of growing human influence with many growing urban areas and second homes and other residential and related development adding to existing land uses and activities such as forestry, mining and recreation. In this region, a reasonable approach might be to consider how much the landscape can be disturbed before there are substantive ecological effects. This contrasts with the How Much approach used in settled landscapes that considers the minimum natural land cover that may be required to conserve biodiversity. This different approach was outlined in the unpublished report How Much Disturbance is too Much?: Habitat Conservation Guidance for the Southern Canadian Shield available from Environment Canada (Beacon Environmental 2012). The report does not provide specific guidelines; rather, it assesses the available science and provides an initial review of the topic along with the introduction of general principles and concepts. As a foundation, How Much Disturbance is too Much? stresses the importance of identifying local and regional habitat mosaics. In these defined areas, restricting most types of development should be considered. These areas would consist largely of forested areas with components of diverse wetlands (e.g., swamps and marshes, as well as fens) interspersed with open, shrub and treed rock barrens. These areas could be based on landscapes having relatively high levels of the following characteristics: diversity, naturalness and habitat extent. They could be identified at two levels: (1) Regional Habitat Mosaics and (2) Local Habitat Mosaics. Regional Habitat Mosaics could include blocks of habitat identified as important on a regional scale (e.g., within an ecodistrict or ecoregion) and would likely include many of the Crown Lands in the southern Canadian Shield. Local Habitat Mosaics could include blocks of habitat identified as important on a local scale (e.g., within a county or township) but similarly capture concentrations of diverse natural areas that are largely undisturbed. Local mosaics would contain one or more large blocks of habitat. Planning could be coordinated at both regional and local jurisdictional levels to identify opportunities for ensuring that Regional and Local Habitat Mosaics are complementary or in proximity to each other where possible, and where it makes sense within the given biophysical and land use context. These areas should cover 50 to 60% of a jurisdiction and be connected to one another. Disturbances within the mosaics should be avoided, especially disturbances created by roads of any kind. 12 How Much Habitat is Enough?

15 Table 1. Summary of Wetland, Riparian, Forest and Grassland Habitat Guidelines Parameter Percent wetlands in the watershed and subwatersheds Wetland location in the watershed Guideline Wetland Habitat Ensure no net loss of wetland area, and focus on maintaining and restoring wetland functions at a watershed and subwatershed scale based on historic reference conditions. At a minimum, the greater of (a) 10% of each major watershed and 6% of each subwatershed, or (b) 40% of the historic watershed wetland coverage, should be protected and restored. Wetlands can provide benefits anywhere in a watershed, but particular wetland functions can be achieved by rehabilitating wetlands in key locations, such as headwater areas (for groundwater discharge and recharge), floodplains and coastal wetlands. Consideration should also be given to protecting networks of isolated wetlands in both urban and rural settings. Critical Function Zones should be established around wetlands based on knowledge of species present and their use of habitat types. Amount of natural vegetation adjacent to the wetland Wetland proximity Wetland area, shape and diversity Protection Zones should protect the wetland attributes from stressors. Recommended widths should consider sensitivities of the wetland and the species that depend upon it, as well as local environmental conditions (e.g., slopes, soils and drainage), vegetative structure of the Protection Zone, and nature of the changes in adjacent land uses. Stressors need to be identified and mitigated through Protection Zone design. Wetlands that are in close proximity to each other, based on their functions, or that are in close proximity to other natural features, should be given a high priority in terms of landscape planning. Capture the full range of wetland types, areas and hydroperiods that occurred historically within the watershed. Swamps and marshes of sufficient size to support habitat heterogeneity are particularly important, as are extensive swamps with minimum edge and maximum interior habitat to support area-sensitive species. Focus on restoring marshes and swamps. Restore fens under certain conditions. Wetland restoration Width of natural vegetation adjacent to stream Percent of stream length naturally vegetated For effective restoration, consider local site conditions, have local sources to propagate new vegetation, and wherever possible refer to historic wetland locations or conditions. Prioritize headwater areas, floodplains and coastal wetlands as restoration locations. Riparian Habitat Both sides of streams should have a minimum 30-metre-wide naturally vegetated riparian area to provide and protect aquatic habitat. The provision of highly functional wildlife habitat may require total vegetated riparian widths greater than 30 metres. 75% of stream length should be naturally vegetated. How Much Habitat is Enough? 13

16 Parameter Percent of an urbanizing watershed that is impervious Percent forest cover Area of largest forest patch Forest shape Percent of watershed that is forest cover 100 m from forest edge Proximity to other forested patches Fragmented landscapes and the role of corridors Guideline Urbanizing watersheds should maintain less than 10% impervious land cover in order to preserve the abundance and biodiversity of aquatic species. Significant impairment in stream water quality and quantity is highly likely above 10% impervious land cover and can often begin before this threshold is reached. In urban systems that are already degraded, a second threshold is likely reached at the 25 to 30% level. Forest Habitat 30% forest cover at the watershed scale is the minimum forest cover threshold. This equates to a high-risk approach that may only support less than one half of the potential species richness, and marginally healthy aquatic systems; 40% forest cover at the watershed scale equates to a medium-risk approach that is is likely to support more than one half of the potential species richness, and moderately healthy aquatic systems; 50% forest cover or more at the watershed scale equates to a low-risk approach that is likely to support most of the potential species, and healthy aquatic systems. A watershed or other land unit should have at least one, and preferably several, 200-hectare forest patches (measured as forest area that is more than 100 metres from an edge). To be of maximum use to species such as forest breeding birds that are intolerant of edge habitat, forest patches should be circular or square in shape. The proportion of the watershed that is forest cover and 100 metres or further from the forest edge should be greater than 10%. To be of maximum use to species such as forest birds and other wildlife that require large areas of forest habitat, forest patches should be within two kilometres of one another or other supporting habitat features. Big Woods areas, representing concentrations of smaller forest patches as well as larger forest patches, should be a cornerstone of protection and enhancement within each watershed or land unit. Connectivity width will vary depending on the objectives of the project and the attributes of the forest nodes that will be connected. Corridors designed to facilitate species movement should be a minimum of 50 to 100 metres in width. Corridors designed to accommodate breeding habitat for specialist species need to meet the habitat requirements of those target species and account for the effects of the intervening lands (the matrix). Forest quality species composition and age structure Watershed forest cover should be representative of the full diversity of naturally occurring forest communities found within the ecoregion. This should include components of mature and old growth forest. 14 How Much Habitat is Enough?

17 Parameter Where to protect and restore Habitat type and area Landscape configuration, heterogeneity and connectivity Patch size Landscape heterogeneity Guideline Grassland Habitats Focus on restoring and creating grassland habitat in existing and potential grassland landscapes. Maintain, restore and create native grassland patches to their historic extent and type at a county, municipal and/or watershed scale considering past presence and current conditions. Grassland habitat patches should be clustered or aggregated, and any intervening land cover should be open or semi-open in order to be permeable to species movement. Maintain and create small and large grassland patches in existing and potential local grassland landscapes, with an average grassland patch area of greater than or equal to 50 hectares and at least one 100-hectare patch. Some grassland habitat should be located adjacent to hedgerows, riparian and wetland habitats for species that require different habitat types in close proximity. How Much Habitat is Enough? 15

18 2. Habitat Guidelines 2.1 Wetland Habitat Guidelines Wetlands are defined here following the Ontario Wetland Evaluation System of the Ontario Ministry of Natural Resources. Essentially, these are areas where hydrophytic (water-adapted) plants comprise 50% or more, by cover, of the vegetation in a given area, and when standing water is present, it is less than 2 metres deep. Wetlands can provide valuable ecological and hydrological functions at both site-specific and watershed scales. Many of southern Ontario s flora and fauna inhabit wetlands during part or all of their life cycle, including many species at risk. Wetlands are known to be biologically diverse habitats, tending to support a wider range of flora and fauna than either temperate upland forests or grasslands, particularly on a species per area basis (Comer et al. 2005; Gibbons et al. 2006; Meyer et al. 2003). Wetland types typically have acidic organic soils, and often contain Sphagnum mosses and ericaceous shrubs (Ericaceae, a family of plants commonly found in acidic and infertile conditions). Fens are peat-accumulating wetlands with groundwater as the dominant water source, which support a variety of plant species, including orchids, sedges and grasses. Swamps are wetlands dominated by trees and shrubs, with periodic standing water, limited drainage, and often neutral or slightly acidic organic soils. Marshes are wetlands that are almost always flooded and are characterized by a mixture of emergent, floating and submerged aquatic vegetation such as reeds, sedges, pondweeds and water lilies. Understanding the diversity of different wetlands is important for conservation planning and restoration because of their differing hydrologic regimes and vegetative structures, each supporting unique assemblages of species and combinations of ecological functions. The wetland classification system for southern Ontario, which is the standard that has been in place since the 1980s, divides wetlands in the Mixedwood Plains into four types: bogs, fens, swamps and marshes. Bogs are peat-accumulating wetlands that trap precipitation as the major water source. They Figure 5. Derryville Bog, Durham Region, Ontario: one of very few true bogs in southern Ontario Beacon Environmental Bogs and fens Bogs are highly specialized environments, and true bogs are rare in the southern part of the 16 How Much Habitat is Enough?

19 Great Lakes basin. They receive almost all of their water and nutrients from precipitation, are acidic, have very low productivity and are dominated by plants that are adapted to low nutrient levels. Fens receive most of their water and nutrients from groundwater, and they may be either nutrient-rich or nutrient-poor, with nutrient-rich fens having a greater diversity of species while the nutrient-poor share many characteristics with bogs. While they play major roles in the hydrology and ecology of the Boreal Shield Hudson Plains Ecozone, bogs and nutrient-poor fens are relatively rare and unique in the south and play less of a role in providing wildlife habitat than the larger, more extensive, and more diverse swamps and marshes of the Mixedwood Plains. seasonal habitat attributes, with the understorey being dominated by wetland species early in the growing season, and plants adapted to drier conditions playing a greater role later in the year. In addition, spring flooding creates ephemeral ponds that are used for breeding by frogs, toads and salamanders. These same pools are also important breeding areas for invertebrates such as some caddisflies and midges, and these, in turn, are important food for bats and many bird species. Swamps Swamps have more than 25% tree or shrub cover; swamps are a major natural landscape feature because of their sheer extent (they comprise almost 90% of the remaining wetland area in southern Ontario). Swamps make significant contributions to forest cover and to watershed hydrology and aquatic health. They also disproportionately contribute to biodiversity because of the variety of cover types they provide. This includes providing mammal wintering areas, sources for a high proportion of cold-water streams, habitat for forest-interior or area-sensitive species, and habitat for many wildlife species, including numerous species at risk, often all simultaneously. The diversity of swamps goes beyond their extent and cover types, and includes a temporal element that is key to the life cycle of many wildlife species. The plant communities of many swamps are dynamic and provide a variety of Figure 6. Flooded swamp at Minesing wetland Graham Bryan Hydrologically, swamps shape the characteristics and health of lower-order streams; they moderate stream hydrology and help maintain water quality. This affects wildlife habitat all the way downstream to the receiving waters of species-rich coastal wetlands. Marshes Marshes are typically associated with the word wetland ; however, they represent only about 10% of the area of wetlands in southern Ontario, and about 5% of all of the province s wetlands (Riley 1988). Today, extensive marshes are rare in the landscape relative to historic conditions, so species that require this habitat are also restricted in their distribution. Several fish How Much Habitat is Enough? 17

20 and wildlife species are totally dependent on marshes, and a high proportion of these are of provincial and federal concern. Some examples of obligate marsh species include Spotted Gar, Spotted Sucker, Banded Killifish, Bullfrog, Pied-billed Grebe, Red-necked Grebe, Least Bittern, Ruddy Duck, King Rail, Virginia Rail, Sora, Common Gallinule, American Coot, Forster s Tern, Black Tern, Marsh Wren and Muskrat. The noteworthy ecological role of marshes is illustrated by Jude and Pappas (1992), who found that of 113 fish species occurring in the Great Lakes, 41.6% were coastal marsh species and 31% used coastal marshes for nursery habitat or feeding. In Lake Ontario marshes, 63.9% of species present used marshes for spawning and 86% of species used marshes as nursery habitat. The importance of marshes to the fish of the Great Lakes and also inland water bodies cannot be overemphasized. Approximately 90% of the fish biomass in Lake Erie is forage fish (also called prey or bait fish), and most of this is produced in wetlands (Keast et al. 1978; Stephenson 1990). Likewise, marshes are key to waterfowl nesting and stopover. resulting in the maintenance of downstream wildlife habitat. Marsh vegetation also stabilizes shorelines and reduces the risk of erosion, protecting and stabilizing in-situ and adjacent habitat, along with reducing sediment delivery to water bodies (Sheldon et al. 2005). Marshes are also the primary building block for wetland restoration; establishing a marsh is often the initial step in the long-term process of establishing swamps and more complex and diverse wetland communities. The following series of wetland habitat guidelines relate to the amount of wetlands and wetland location in a watershed, the amount of vegetation adjacent to a wetland, representation of wetland area and type, and wetland restoration, as summarized in Table 2. As with swamps, inland marshes help shape stream hydrology and aquatic ecosystem health, including playing a major role in improving and maintaining water quality, Figure 7. Long Point a Great Lakes Coastal Marsh Environment Canada Wet meadow marshes a specialized wetland type Wet meadows provide habitat for a high diversity of plants and wildlife in southern Ontario. Rare species known to inhabit wet meadows include Henslow s Sparrow and Yellow Rail. Wet meadows are habitats subject to temporary flooding dominated by herbaceous plants typical of moist soils, and can be easily overlooked as upland habitats if seen during a dry period. These types of wetlands occur predominantly along shorelines of large lakes and rivers where invasion by woody species is also prevented by ice scouring and waves. 18 How Much Habitat is Enough?

21 Table 2. Summary of Wetland Habitat Guidelines Parameter Percent wetlands in the watershed and subwatersheds Guideline Ensure no net loss of wetland area, and focus on maintaining and restoring wetland functions at a watershed and subwatershed scale based on historic reference conditions. At a minimum, the greater of (a) 10% of each major watershed and 6% of each subwatershed, or (b) 40% of the historic watershed wetland coverage, should be protected and restored. Wetland location in the watershed Wetlands can provide benefits anywhere in a watershed, but particular wetland functions can be achieved by rehabilitating wetlands in key locations, such as headwater areas (for groundwater discharge and recharge), floodplains and coastal wetlands. Consideration should also be given to protecting networks of isolated wetlands in both urban and rural settings. Critical Function Zones should be established around wetlands based on knowledge of species present and their use of habitat types. Amount of natural vegetation adjacent to the wetland Protection Zones should protect the wetland attributes from stressors. Recommended widths should consider sensitivities of the wetland and the species that depend upon it, as well as local environmental conditions (e.g., slopes, soils and drainage), vegetative structure of the Protection Zone, and nature of the changes in adjacent land uses. Stressors need to be identified and mitigated through Protection Zone design. Wetland proximity Wetland area, shape and diversity Wetlands that are in close proximity to each other, based on their functions, or that are in close proximity to other natural features, should be given a high priority in terms of landscape planning. Capture the full range of wetland types, areas and hydroperiods that occurred historically within the watershed. Swamps and marshes of sufficient size to support habitat heterogeneity are particularly important, as are extensive swamps with minimum edge and maximum interior habitat to support areasensitive species. Focus on restoring marshes and swamps. Restore fens under certain conditions. Wetland restoration For effective restoration, consider local site conditions, have local sources to propagate new vegetation, and wherever possible refer to historic wetland locations or conditions. Prioritize headwater areas, floodplains and coastal wetlands as restoration locations. How Much Habitat is Enough? 19

22 2.1.1 Percent Wetlands in the Watershed and Subwatersheds > Guideline Ensure no net loss of wetland area, and focus on maintaining and restoring wetland functions at a watershed and subwatershed scale based on historic reference conditions. At a minimum, the greater of (a) 10% of each major watershed and 6% of each subwatershed, or (b) 40% of the historic watershed wetland coverage, should be protected and restored. > Rationale All watersheds in southern Ontario currently have less wetland cover than they did prior to extensive European settlement (c. 1800), with losses exceeding 70% in many jurisdictions (Ducks Unlimited Canada 2010; Snell 1987). The guideline addresses basic minimal generic ecological functions and does not address the overall loss of unique wetland ecosystems that dominated portions of southern Ontario. Any investment in wetland restoration beyond the minimum guideline toward the historical wetland baseline will result in enhanced wetland functions and contributions to ecological goods and services. Maintenance of wetland cover across a watershed provides many ecological and hydrologic benefits. The extent of these benefits varies depending on a variety of biophysical factors including predominant landforms and soils, wetland locations, types of wetland, and predominant land uses (Flanaghan and Richardson 2010; Keddy 2010; Zedler 2003). Historically the overall wetland coverage within the Great Lakes basin exceeded 10% (Detenbeck et al. 1999), but there was significant variability among watersheds and jurisdictions. For example, recent studies of three watersheds in the Lake Simcoe Watershed (i.e., Whites Creek, Maskinonge River and Innisfil Creek) have resulted in wetland cover estimates of 68, 52 and 59% respectively (A. Norman, OMNR, London, pers. comm. 2011), while analyses by Ducks Unlimited Canada (2010) indicate that presettlement wetland cover in some counties such as Peel was as low as 7.6%, and as high as 83.4% in others such as Essex. Using historical reference points can be useful for helping to determine an appropriate level of wetland cover at the watershed or jurisdictional scale (e.g., Bedford 1999; Puric-Mladenovic and Strobl 2006). More recent historical mapping (c ) can also be used to help target former wetland areas that may be suitable for restoration. One of the real challenges for maintaining key ecological and hydrological functions is in estimating the critical threshold for wetland cover in a watershed (or jurisdiction). Not surprisingly, the science on this topic shows variation among watersheds, and focuses on protection of hydrologic functions. In Wisconsin, Hey and Wickencamp (1996) examined nine watersheds and found that increasing the amount of wetland in a watershed resulted in reduced yields of water downstream, reduced flooding, higher base flows and reduced occurrence of high flows. However, these responses flattened very rapidly 20 How Much Habitat is Enough?

23 beyond 10% wetland cover. Zedler (2003) determined that flood abatement, water quality improvement and biodiversity support declined significantly when about 60% of the Upper Midwest s historical wetland area was drained, suggesting retention of about 40% of that area s wetland cover would continue to support those key functions. A study conducted by the University of Minnesota (Johnston et al. 1990) found that watersheds in the southern United States containing less than 10% wetlands were more susceptible to incremental losses of wetland functions than watersheds with more wetlands. This condition was found to be particularly true for flood control and suspended solids loadings. These studies support the importance of considering watershed-specific differences in historical wetland cover, information that is readily available to jurisdictions in southern Ontario, in older as well as in updated reports (i.e., Ducks Unlimited Canada 2010; Snell 1987). In applying this guideline, current wetland cover, topography, soils and extent of impervious surfaces in a specific watershed must also be considered. In the absence of such information, the guideline of 10% cover at the watershed and 6% cover at the subwatershed scales can be used to ensure that a minimum level of wetlands are distributed throughout the watershed. While the maintenance of wetland functions is as or more important than the maintenance of wetland area, given the limited nature of our current understanding of wetland functions, particularly at the watershed scale, wetland area serves as a useful surrogate measure. In many watersheds in southern Ontario, particularly urbanized watersheds, it is not possible to return to historical or preurbanization levels of wetland cover or function because of the degree and nature of alteration that has already occurred. Nonetheless, given the known extent of wetland losses in this part of the province and the critical hydrological and ecological functions provided by these habitats, a no net loss approach combined with a commitment to work towards at least 40% of historical levels of coverage where it does not already exist is recommended to yield tangible benefits for communities and wildlife. The guideline can be achieved, in order of priority, through: (1) protection of extant wetlands; (2) enhancement of extant wetlands; (3) restoration of wetlands in historical locations; and (4) creation of wetlands in suitable areas. How Much Habitat is Enough? 21

24 2.1.2 Wetland Location > Guideline Wetlands can provide benefits anywhere in a watershed, but particular wetland functions can be achieved by rehabilitating wetlands in key locations, such as headwater areas (for groundwater discharge and recharge), floodplains and coastal wetlands. Consideration should also be given to protecting networks of isolated wetlands in both urban and rural settings. > Rationale Wetlands anywhere within a watershed will provide both ecological and hydrological benefits, but are increasingly being understood to provide different functions depending on their location in the watershed, as well as the characteristics of the watershed itself. The extent to which wetlands can provide water quality benefits has been linked to the biophysical characteristics of the watershed (Norton and Fisher 2000) as well as the land use context (Mitsch and Gosselink 2000; Zedler 2003). Model-based analyses indicate that water quantity and quality benefits are derived from protection of a range of wetland sizes throughout the watershed, with small (i.e., 0.2 hectares) headwater wetlands being important for sediment removal, medium-sized (i.e., 10 hectares) mid-reach wetlands retaining significant amounts of phosphorus, and large (i.e., 250 hectares) floodplain wetlands effectively storing and attenuating long-period hydrologic flows (Cohen and Brown 2007). Earlier literature has also shown that wetlands can perform different functions depending on flow levels. For example, Johnston et al. (1990) found in their watershed-scale study in Minnesota that wetlands adjacent to streams were more effective at attenuating suspended solids, total phosphorus and ammonia during periods of high flow, and more effective at removing nitrates in periods of low flow. These results support the need for wetlands of various sizes and in various locations in the watershed, at least for provision of a more full range of water quality benefits. In headwater areas, wetlands can provide beneficial functions. For swamps, these include protection of the quality of groundwater (discharge or recharge or both), introduction of leaves and woody debris that are essential to providing habitat for fish and macroinvertebrates downstream (Gurnell et al cited in Detenbeck et al. 1999), and reducing the warming of streams at the source. In turn, good water-quality conditions in higher portions of watersheds are likely to benefit downstream coastal wetland ecosystems (e.g., Crosbie and Chow-Fraser 1999). Janisch et al. (2011) speak to the benefits of headwater wetlands in influencing headwater surface area processes, including improving resilience of streams following disturbances. Further downstream, palustrine and riverine wetlands are important in reducing and asynchronizing peak flows, improving water quality, and providing habitat for aquatic invertebrates, fish and other wildlife. Richardson et al. (2011) were able to demonstrate, through a multi-phased restoration project of a stream and riparian 22 How Much Habitat is Enough?

25 wetland area located within the lower portion of a watershed in North Carolina, that wetlands in lower reaches associated with riparian areas can also provide significant benefits. They documented reduced downstream water pulses, nutrients, coliform bacteria and stream erosion; a substantial attenuation of nitrogen and phosphorus; and an increase in wetland plant abundance and diversity within the floodplain. In coastal areas such as the Great Lakes, marshes are crucial habitat for fish. Wetland habitats in lakes tend to support more fish biomass than open parts of the lake, and are important in supporting fish production and species diversity (Petzold 1996; Trebitz et al. 2009). Petzold (1996) found about 60% of Lake Superior s fish biomass was associated with wetlands and concluded that these habitats are critical for the fisheries of the entire lake. This is supported by observations that changes in the amount and type of wetlands at Long Point have affected the fish assemblages populating all of Lake Erie (T. Whillans, Trent University, Peterborough, pers. comm. 2011). Recent literature indicates that geographically isolated wetlands also provide water quality and flow regulation services, as well as maintenance of amphibian and reptile biodiversity (Comer et al. 2005). A study undertaken by Russell et al. (2002) in managed young-growth forests in the Coastal Plain of South Carolina found that isolated wetlands were focal points of amphibian and reptile richness and abundance and contributed more to regional biodiversity than would be expected based on their size and ephemeral hydrology. McKinney and Charpentier (2009) found that geographically isolated wetlands contribute to stormwater retention, flood prevention and maintenance of water quality. Wetlands can also provide benefits that address specific objectives or problems at a watershed or more site-specific scale. Wetlands located within urban or agricultural settings act to improve water quality by retaining nutrients and sediments, and providing stormwater management (Flanagan and Richardson 2010; McKinney and Charpentier 2009). Flanagan and Richardson (2010) found that former wetland areas converted to agricultural uses were linked to higher levels of phosphorus in nearby water bodies and recommended restoration of at least some of these areas to wetland to address this issue. Some of these more localized benefits are discussed in the following section. How Much Habitat is Enough? 23

26 2.1.3 Amount of Natural Vegetation Adjacent to the Wetland > Guideline Critical Function Zones should be established around wetlands based on knowledge of species present and their use of habitat types. Protection Zones should protect the wetland attributes from stressors. Recommended widths should consider sensitivities of the wetland and the species that depend upon it, as well as local environmental conditions (e.g., slopes, soils and drainage), vegetative structure of the Protection Zone, and nature of the changes in adjacent land uses. Stressors need to be identified and mitigated through Protection Zone design. > Rationale The amount of natural habitat that is located adjacent to wetlands can be important to the maintenance of wetland functions and attributes, particularly for wetland-dependent species that rely on these adjacent natural areas for portions of their life cycle. In cases where these adjacent natural areas form an intrinsic part of the wetland ecosystem, providing a variety of habitat functions for wetlandassociated fauna that extend beyond the wetland limit, these lands can be described as Critical Function Zones (CFZs). Beyond habitat functions, adjacent natural and semi-natural areas can also provide what are often called buffer functions by protecting the wetland (and its associated CFZs) from external stressors. These stressors are typically associated with human-induced changes in land use and include sedimentation, contaminants, noise, light, physical disturbances (e.g., trampling or garbage dumping), and the introduction and spread of invasive species. These adjacent areas that serve primarily a protective function are best described as Protection Zones (PZs). Determining the appropriate amount of natural area adjacent to a wetland requires independent consideration of the CFZ and the PZ, and the functions of the two should not be confused. 24 How Much Habitat is Enough?

27 Critical Function Zones and Protection Zones defined The term Critical Function Zone (CFZ) describes non-wetland areas within which biophysical functions or attributes directly related to the wetland occur. This could, for example, be adjacent upland grassland nesting habitat for waterfowl (that use the wetland to raise their broods). The CFZ could also encompass upland nesting habitat for turtles that otherwise occupy the wetland, foraging areas for frogs and dragonflies, or nesting habitat for birds that straddle the wetland-upland ecozone (e.g., Yellow Warbler). A groundwater recharge area that is important for the function of a wetland but located in the adjacent lands could also be considered part of the CFZ. Effectively, the CFZ is a functional extension of the wetland into the upland. It is not a buffer for the wetland. See Figure 8, The Critical Function and Protection Zones. Once identified, the CFZ (along with the wetland itself) needs to be protected from adverse effects that originate from external sources by a Protection Zone (PZ). The PZ s primary function is to protect the wetland and its associated functions from stressors associated with activities in, or changes to, the lands external to the wetland. The PZ acts as a buffer in response to stressors on wetland water Figure 8. The Critical Function and Protection Zones quality, water quantity (including the timing and degree of changes in water levels), habitat functions, or all three. PZs are analogous to filter strips and are typically vegetated areas for intercepting stormwater runoff and attenuating and transforming associated nutrients or other contaminants. They also provide physical separation from one or more stressors such as noises or visual disturbances. And they protect against direct human-associated intrusions into the wetland. Such functions are well-established in the literature (e.g., Kennedy et al. 2003; Lowrance et al. 2002; Passeport et al. 2010; Sheldon et al. 2005; Thompson et al. 2004; Woodard and Rock 1995). Depending on the nature of the stressors and the sensitivities of the wetland, alternative PZ design features such as a fence can be effective. Fundamentally, the PZ is aimed at reducing impacts on wetland functions that originate from the upland side. Critical Function Zone Dragonfly occurrence 160 m Wetland Protection Zone Waterfowl nesting 200 m Painted turtle nesting m How Much Habitat is Enough? 25

28 A word about buffers and adjacent land The term buffer is a common term used for lands adjacent to wetlands and watercourses. In How Much Habitat is Enough?, the term adjacent lands is commonly used and is meant to be interpreted literally as the lands immediately adjacent to a wetland. This is because adjacent lands may have buffer functions, important non-buffer habitat functions, or both. These distinctions are important. For example, an area that serves a CFZ for one species may also act as a buffering PZ for another species CFZ, or as a wetland or stream buffer. In land use planning in Ontario, the term adjacent lands is used specifically to refer to those lands contiguous to a specific natural heritage feature or area where it is likely that development or site alteration would have a negative impact on the feature or area (MMAH 2005). These are typically prescribed as a set distance (e.g., between 50 and 120 m) from the boundary of the natural feature. The combined CFZ and its PZ may range in total width from a few metres to hundreds of metres. Management objectives, individual characteristics of the wetland, ecological interactions with upland areas, and the source, magnitude and frequency of potential stressors and engineering options all contribute to the design of effective CFZ and PZ areas. For CFZ determination, a good understanding of the local biophysical context, hydrologic regime and the species using the given wetland, as well as the nature and extent of their nonwetland habitat requirements, is required (e.g., Guerry and Hunter 2002; Pope et al. 2000). Guidance regarding non-wetland habitat requirements for wildlife is increasingly available, and some of this current information is summarized in Table 3. For wildlife, the variability in ranges of CFZs is great because of both inter- and intra-species variability in documented distances travelled for feeding and overwintering, as well as variability among breeding sites (e.g., Nichols et al. 2008). In time, more data will become available as further research (e.g., ongoing radio-telemetry studies) will lead to a more complete understanding of various species habitat requirements. 26 How Much Habitat is Enough?

29 Table 3. Selected Critical Function Zone Data for Wetlands Species Midland Painted Turtle Extent of Critical Function Zone from Wetland Edge Reptiles Maximum 600 m, mean 60 m; range 1 to 164 m; mean 90 m, range 1 to 621 m 1 to 620 m, mean 90 m for nesting Reference Semlitsch and Bodie 2003 Christens and Bider 1987 Notes Review of three studies 85 m for nesting, 54 m for dormancy Joyal et al Distances are mean plus standard deviation Spotted Turtle 75 to 312 m for nesting; dormancy up to 412 m Milam and Melvin 2001 Maximum 150 m; range 3 to 265 m; range 60 to 250 m Semlitsch and Bodie 2003 Review of three studies Blanding s Turtle 380 m for nesting, 18 m for basking and 114 m for dormancy Mean 815 m, range 650 to 900 m; mean 135 m, range 2 to 1115 m; mean 168 m Joyal et al Semlitsch and Bodie 2003 Distances are mean plus standard deviation Review of three studies Spiny Softshell Mean 3 m; range 2 to 3 m; mean 0.3 m; mean 5 m Semlitsch and Bodie 2003 Review of four studies; latter two studies single individuals Snapping Turtle Mean 94 m, range 38 to 141 m; mean 37 m, range 1 to 183 m; mode 25 m, maximum 100 m; mean 27 m, range 1 to 89 m Semlitsch and Bodie 2003 Review of four studies Wood Turtle Mean 27 m, range 0 to 500 m; maximum 600 m; mean 60 m, maximum 200 m Semlitsch and Bodie 2003 Review of four studies Northern Map Turtle Mean 2 m, range 2 to 3 m Semlitsch and Bodie 2003 Review of one study Eastern Musk Turtle Mean 7 m, range 3 to 11 m Semlitsch and Bodie 2003 Review of one study Northern Watersnake No adjacent lands area recommended Maximum 6 m Attum et al Semlitsch and Bodie 2003 Species fairly sedentary and generally does not require upland habitat Review of one study How Much Habitat is Enough? 27

30 Species Extent of Critical Function Zone from Wetland Edge Reference Notes Frogs Western Chorus Frog Maximum 213 m, mean 75 m Semlitsch and Bodie 2003 Review of one study Post-breeding movements ranged from 102 m to 340 m, median 169 m Baldwin et al. 2006a Recommends conservation of network connected habitats 30 m inadequate to support viable populations Harper et al Did not test beyond 30 m Wood Frog 11 to 35 m partially mitigates timber harvest impacts Perkins and Hunter 2006 Study is in forested landscape where the nonbuffered area is logged 40% of individuals wintered further than 100 m from breeding pond Regosin et al Recommends maintenance of suitable terrestrial habitat beyond 100 m Mean 36 ± 25 m for foraging Lamoureux et al Green Frog Mean 137 m, maximum 457 m; mean 121 m, maximum 360 m; mean 485 m, range 321 to 570 m Semlitsch and Bodie 2003 Review of three studies Bullfrog Mean 406 m Salamanders Semlitsch and Bodie 2003 Mean maximum 106 m Veysey et al Review of one study Salamanders used clearcut areas to some degree 30 m inadequate to support viable populations Harper et al Did not test beyond 30 m 60% of individuals wintered further than 100 m from breeding pond Regosin et al Recommend maintenance of suitable terrestrial habitat beyond 100 m Spotted Salamander Mean 67 m, range 26 to 108 m; mean 103 m, range 15 to 200 m; mean 64 m, range 0 to 125 m; mean 150 m, range 6 to 220 m; mean 192 m, range 157 to 249 m, mean 118 m, range 15 to 210 m Semlitsch and Bodie 2003 Review of six studies Range 3 to 219 m, mean m for overwintering Faccio 2003 Recommended life zone to encompass 95% of population was 175 m 28 How Much Habitat is Enough?

31 Species Ambystoma salamanders Jefferson Salamander Blue-spotted Salamander Eastern Newt Various species in Ontario Extent of Critical Function Zone from Wetland Edge Mean 125 m for adults, 70 m for juveniles Range 3 to 219 m, mean m for overwintering Mean 39 m, range 22 to 108 m; mean 92 m, range 15 to 231 m; mean 252 m, range 20 to 625 m: mean 250 m 52% of individuals wintered more than 100 m from breeding pond No distances given No distances given Nesting Waterfowl 0 to more than 400 m; 90% were within 200 m. About 20% of nests were inside or within 25 m of wetlands Reference Semlitsch 1998 Faccio 2003 Semlitsch and Bodie 2003 Regosin et al Roe and Grayson 2008 Rinehart et al Henshaw and Leadbeater 1998 Dragonflies and Damselflies Dragonflies, especially Halloween Pennant, found 10 to 160 m from wetland edge Bried and Ervin 2006 Notes Recommended life zone to encompass 90% of population was 164 m Recommended life zone to encompass 95% of population was 175 m Review of four studies Recommend maintenance of suitable terrestrial habitat beyond 100 m Newts are wide-ranging and active in the terrestrial habitat Proximity to developed land cover essentially precluded newt occupancy Based on data from 102 nests at coastal marshes over two years. May be applicable where suitable waterfowl nesting habitat is present. Study in Mississippi. There was a different sex balance at different distances. Various species Bogs with natural habitat around them had a higher dragonfly and damselfly abundance than those with peat mining in adjacent lands Bonifait and Villard 2010 Study in bogs in New Brunswick How Much Habitat is Enough? 29

32 Like CFZs, optimal PZ widths also vary depending on a number of site-specific factors as well as the land use context. Of primary importance is understanding the desired function(s) that the PZ is expected to perform. Key parameters that need to be considered in determining PZ widths include local hydrologic dynamics, slope, soil type(s), the vegetative composition of the buffer, and the extent and nature of the anticipated stressors (Ducros and Joyce 2003; Hawes and Smith 2005; Johnson and Buffler 2008; Polyakov et al. 2005). Examples of recommended PZ (buffer) widths for wetlands are provided in Table 4. PZ width can also vary depending on its anticipated uses. How Much Habitat is Enough? encourages a shift towards the development of multicriteria evaluation approaches for PZs (or buffers) (van der Merwe and Lohrentz 2001). This approach encourages the identification and prioritization of various criteria that are selected on a site-specific basis. This could result, for example, in the encouragement of some land uses or activities within the PZs (e.g., trails), but not within the CFZs. The use of distinct bands within the adjacent lands area can help resolve some land use challenges when urban development is proposed close to wetlands. For example, an appropriately sized and designed PZ can accommodate trails that support opportunities for hiking and cycling, as well as nature interpretation and appreciation, or urban infrastructure such as stormwater management facilities. Based on current knowledge, the literature increasingly indicates that the habitat requirements for wildlife tend to result in the widest and most varied CFZs (e.g., in the order of 100 metres or more). In contrast, maintaining water quality and aquatic habitat functions in streams and wetlands can often be achieved with zones of approximately only 30 metres (although there is a fair amount of site-specific variability). There are no known studies that actually test the ability of different buffer types or widths to protect wetland habitats (whether for plants or wildlife or both). Therefore, PZ recommendations related to wildlife, where provided, are typically extrapolated from measurements of impacts to various wetland species. Such recommendations may overestimate or underestimate the actual buffer widths required, and more research is required to address this knowledge gap. 30 How Much Habitat is Enough?

33 Table 4. Examples of Recommended Protection Zones or Buffers to Wetlands Stressor Sediment Herbicide drift from agricultural lands Non-point source agricultural pollutants Suggested Extent of Protective Zone Reference 6 m Hook to 60 m Skagen et al Strip at edge of cultivated fields (data indicate > 6 to 9 m) 16.3 m grass/woody strip (riparian) Boutin and Jobin 1998 Lee et al Notes High attenuation rate regardless of slope (0 to 20%) Range based on literature review Cites other studies suggesting 5 to 10 m Removed > 97% of sediment, narrower (7 m) grass provided some benefits Residential stormwater 15 m; 23 to 30 m on slopes greater than 12% Woodard and Rock 1995 Groundcover type also very important Human disturbance, landscaping (e.g., wood piles, composting) 19 to 38 m Matlack 1993 Fencing may achieve same results in less width Nitrate 16 to 104 m Basnyat et al Human disturbance by watercraft Human disturbance, recreation-related (e.g., camping, hacked trees) Human disturbance (on nesting Great Blue Herons) Urban cats More than 80 m Rodgers and Schwikert to 130 m Matlack m 190 m Erwin 1989; Rodgers and Smith 1995 Haspel and Calhoon 1991 Objective was > 90% nitrate removal Based on a flush distances* of approximately 45 to 80 m for Great Lakes species (no waterfowl) Flush distance* was 32 m ± 5.5 m; 40 m added to mitigate antagonistic behaviour Measured distance predation rates on wildlife extended into adjacent natural area * Flush distance = proximity of disturbance that will cause bird to leave nest How Much Habitat is Enough? 31

34 Useful guidance is also available from review papers that summarize data from a wide range of sources. Several review papers that have examined the available data on recommended adjacent natural areas for wetlands are summarized in Table 5. Notably, these reviews tend not to distinguish between CFZs and PZs, and instead include all adjacent land requirements into the category of buffers. The Planner s Guide to Wetland Buffers for Local Governments (Nichols et al. 2008) concludes that, on average, between 30 and 91 metres of adjacent natural areas are required to support wildlife habitat functions of wetland-dependent species (with some studies recommending larger widths). The comprehensive review of temperate freshwater wetlands by Sheldon et al. (2005) concludes that there is no one effective buffer width, and recommends three possible ranges for both wildlife functions (i.e., CFZs) and buffers (i.e., PZs) combined depending on (a) the wetland s level of habitat function, and (b) the intensity of adjacent land uses ranging from 8 metres to more than 92 metres. Table 5. Selected Reviews or Guidelines that Consider Areas of Land Adjacent to Wetlands Reference Brown et al Castelle et al Lowrance et al Norman 1996 Kennedy et al Recommendations for Adjacent Lands* 30 to 168 m for groundwater protection; 23 to 114 m for sedimentation; 98 to 223 m for wildlife habitat Minimum of 15 to 30 m under most circumstances, but sitespecific Range from 1 to 30 m Baseline adjacent lands area of 50 m, then subject to site-specific considerations (e.g., waterfowl production or sensitive hydrology) Recommended minimums: 25 m for nutrient and pollution removal 30 m for microclimate regulation and sediment removal 50 m for detrital input and bank stabilization 100 m for wildlife habitat functions Notes Study in Florida (with a particular geology). Recommendations based on consideration of landscape conditions and information from literature review. Based on U.S. studies. Literature review encompasses sediment removal, nutrient removal, stormwater runoff, moderation of temperature, habitat diversity and habitat protection functions. Based on U.S. studies. Focus on water quality functions (sedimentation and erosion, nutrient management, and pathogens and pesticides). Southern Ontario context. Focus on water quality functions (erosion control and reduced contamination transportation) in agricultural settings. Cautions against very narrow buffer strips. Based on U.S. studies between 1990 and How Much Habitat is Enough?

35 Reference Sheldon et al Bentrup 2008 Nichols et al Recommendations for Adjacent Lands* 8 to 23 m for wetlands with minimal habitat functions and low-intensity land uses adjacent to the wetland 15 to 46 m for wetlands with moderate habitat functions and moderate or highintensity land uses adjacent to the wetland 46 to 92+ m for wetlands with high habitat functions, regardless of the intensity of the land uses adjacent to the wetland No distances recommended 9 m to 30 m for sediment and phosphorus removal 30 m to 49 m for nitrogen removal 30 m to 91 m for wildlife protection (with some studies showing more) Notes Focus on freshwater wetlands in Washington State Synthesis documents generally recommend adjacent land area widths of between 15 to 100 m, but no one width can be recommended Recommended principles: For Critical Function Zones larger species require larger widths, width should increase with length and in human dominated areas, and Critical Function Zones that need to function for a longer time should be wider For Protection Zones width can be variable if variable runoff, but should be wider on steeper slopes and on soils with lower infiltration capacities Review based on 50 U.S. ordinances, hundreds of scientific papers, and analyses of wetland adjacent land area performance * Includes Critical Function Zones and Protection Zones (buffers) Notably, even protection of wetlands and their functions through the identification and implementation of CFZs and PZs will not fully conserve the habitat functions of wetlands on a landscape scale. Implementation of these sitespecific measures must be considered in the broader context of natural heritage protection on a watershed or regional scale. This is well recognized for water quality and quantity (Sheldon et al. 2005), which can be more heavily influenced by changes in land use in the broader landscape than by localized habitat protection efforts, and for wildlife such as amphibians and plants that rely on successful dispersal and migration in the broader landscape for population maintenance (Bauer et al. 2010; Keddy 2010; Semlitsch 2008). How Much Habitat is Enough? 33

36 2.1.4 Wetland Area, Shape and Diversity > Guideline Capture the full range of wetland types, areas and hydroperiods that occurred historically within the watershed. Swamps and marshes of sufficient size to support habitat heterogeneity are particularly important, as are extensive swamps with minimum edge and maximum interior habitat to support area-sensitive species. > Rationale Extensive, heterogeneous wetlands as well as less extensive, isolated wetlands both make significant contributions to supporting biodiversity at the local and watershed scales. The presence of larger, contiguous swamps and marshes (e.g., more than 30 hectares) are important for area-sensitive species such as Prothonatory Warbler and Black Tern. However, the presence of complexes of smaller, more isolated wetlands in the landscape are also important in that they provide habitat for many wetland-dependent amphibians and reptiles. Swamps have the potential to support areasensitive wildlife species (i.e., those that require larger areas of continuous habitat in which to be productive) or edge-intolerant species (i.e., those that prefer to use habitat away from the influence of habitat edges, also sometimes referred to as interior habitat species). In some watersheds with many land use pressures, treed swamps may be the only remaining significant contributors to interior-forest habitat, and the discussion on forest size and species that may be expected (see Section 2.3) applies here as well. However, treed swamps provide interior habitat for a different suite of specialist area-sensitive forest species compared to large patches of upland forest. Larger marshes also have the ability to support area-sensitive wildlife species (Smith and Chow- Fraser 2010). Area-sensitive birds may include species such as Marsh Wren, Black Tern and Forster s Tern. Black Tern will nest in smaller marshes if larger feeding areas are located nearby. Some other species, such as Least Bittern and King Rail, occasionally occur in smaller marshes, but long-term viable populations are associated with extensive marshes. Extensive swamps and marshes also tend to have greater habitat heterogeneity (i.e., the habitat is more varied within them), which in turn tends to support more species of wildlife (e.g., Golet et al. 2001). In marshes, this is called interspersion or the juxtaposition of different marsh communities. High levels of habitat interspersion (e.g., the presence of open water/submerged vegetation, emergent vegetation and in some cases shrubs) within a marsh provide higher-quality habitat for a wider variety of species than, for example, a narrow band of cattails around the shoreline. For example, some species require extensive stands of emergent plants with few or no openings (e.g., Northern Harrier), while others seem to prefer areas dominated by emergent plants but with small, isolated openings (e.g., Least Bittern). The ratio of open water/submerged vegetation to emergent vegetation and the interspersion pattern may 34 How Much Habitat is Enough?

37 vary considerably from year to year because marshes are dynamic systems. However, area remains a key factor, and more extensive marshes are more likely to be used as productive habitat by more species of wildlife (e.g., Attum et al. 2007; Webb et al. 2010). Relatively isolated (i.e., not coastal or riparian), smaller wetlands can also be important for local or regional biodiversity. For example, amphibians such as Wood Frog and Spotted Salamander have been documented in wetlands ranging in size from 0.1 to 5.2 hectares (Babbitt 2005; Lehtinen and Galatowitsch 2001). These wetlands can have variable hydroperiods, may be permanently wet (i.e., year-round) or only seasonally wet (typically in the spring and part of the summer), and can be hot spots for some groups of amphibians, particularly when they do not support predatory or competing fish (Snodgrass et al. 2000; Werner et al. 2007). Interestingly, current research indicates that hydroperiod may be a more important factor than wetland size in determining the diversity that can be supported by isolated wetlands (Baldwin et al. 2006b; Hermann et al. 2005; Paton and Crouch 2002; Snodgrass et al. 2000). Complexes of relatively isolated wetlands also tend to be more supportive of biodiversity than single, isolated ponds. Amphibians and reptiles are known to use multiple local ponds, sometimes over the same season (Joyal et al. 2001; Semlitsch 2008). Waterbirds are also known to use complexes of small wetlands, especially for springtime pairing and feeding, and have been documented using isolated wetlands in urban and peri-urban landscapes in London, Ontario (Pearce et al. 2007). Some birds have specifically adapted to use wetland complexes in the landscape and will readily move between them to forage (e.g., Northern Harriers, herons, dabbling ducks). This is the reason that the Ontario Wetland Evaluation System recognizes the concept of wetland complexes (OMNR 1994). The presence of coarse woody debris in many wetland types particularly swamps, but also riparian areas and terrestrial areas associated with wetland pockets is important to many species. The functions of this debris include providing cover and nutrients for fish and other aquatic organisms, and providing important cover and overwintering habitat for pondbreeding amphibians that spend the bulk of their life cycle in associated uplands (Keddy 2010). Maintenance of the full range of wetland vegetation community types that occur in a watershed is also key to sustaining biodiversity. Meyer et al. (2010) found that at Long Point on Lake Erie, while the overall abundance and diversity of birds was greater in Common Reed habitat, the abundance of marsh-nesting birds was greater in meadow marsh habitat, supporting the need to protect these types of more specialized habitats. The role of wetland shape in supporting habitat and species diversity is difficult to discuss independently because it is so closely related to, and in the literature is often confounded with, habitat area and fragmentation in the landscape (Ewers and Didham 2006). The limited available data indicate that the optimum shape of a wetland varies by wetland type. Swamps, which are a type of forest, are better able to support area-sensitive and edge-intolerant species when they are relatively compact and regularly shaped (e.g., circular or squarish) (see Figure 13, on forest shape determining amount of core habitat, in the Forest Habitat Guidelines). However, some other wetland-dependent species require ecotonal or edge habitat (e.g., How Much Habitat is Enough? 35

38 transitional areas between open water and adjacent uplands) and thrive where there is more edge (Attum et al. 2007; Stevens et al. 2002). Long, narrow marshes may also provide more water quality benefits since they maximize water contact with vegetation that is responsible for the uptake and transformation of many nutrients and other contaminants. The link between wildlife species diversity and abundance, and the presence of wetlands, has been made repeatedly for amphibians. Specific research on hydroperiods of seasonally inundated wetlands in forests has shown that wetlands with longer hydroperiods (but not permanent water) support a higher diversity of amphibians, irrespective of wetland size (Babbitt 2005; Baldwin et al. 2006b; Herrmann et al. 2005). 36 How Much Habitat is Enough?

39 2.1.5 Wetland Proximity > Guideline Wetlands that are in close proximity to each other, based on their functions, or that are in close proximity to other natural features, should be given a high priority in terms of landscape planning. > Rationale Fragmentation of wetland habitats degrades their functions by reducing habitat for species that are less tolerant of disturbances, that require more contiguous habitat, or both, compromising the ability of individuals of a species to effectively disperse and mate with individuals from other populations, and increasing habitat for opportunistic species (such as exotic invasive species and pests). Some of these negative impacts of fragmentation can be offset, at least for some species, by maintaining concentrations of natural habitat fragments within relatively close proximity in a given landscape. This approach is wellrecognized through the approach of the Ontario Ministry of Natural Resources to complexing wetlands that are within 750 metres distance of each other. The benefits of this type of land use planning may be further enhanced by minimizing the scale and extent of built-up land uses (e.g., road size and density) in the lands within such areas. Fragmentation of marshes within lakes can result in depletion of zooplankton and the fish species that depend on them. Even in systems where zooplankton is not a concern, small marsh patches may be ecological traps. They attract fry of many fish species as nursery habitat, but predation rates by common piscivorous (fish eating) fish such as Rock Bass may be very high. Small marshes especially high concentrations of small marshes in a landscape have traditionally been conserved and restored for waterfowl production. Increasingly, the importance of adjacent natural areas, as well as proximity between patches of wetland, has become recognized for a number of other wildlife species. Attum et al. (2008) in a study conducted in Ohio and Michigan found that both Copper-bellied Watersnake and Blanding s Turtle were more likely to occur in wetlands with more surrounding forest, while Attum et al. (2007) conclude that protection of wetland complexes with a range of wetland patch sizes is needed to support Northern Watersnake foraging and mating habits. Stevens et al. (2002) found that occurrence of calling Green Frogs in restored wetlands was positively correlated with proximity to other wetlands (but provided no specific distances), while Houlahan and Findlay (2003) found in their study of 74 Ontario wetlands that amphibian richness was positively correlated with forest cover in adjacent lands, and also identified this trend in previous studies. In terms of numerical distances for proximity, examples from the current literature are summarized in Table 6, and show a great range. How Much Habitat is Enough? 37

40 Table 6. Examples of Distances in Which Positive Relationships to Other Wetlands or Other Habitats Was Documented Wildlife Group or Species Distance Within Which a Positive Response Was Documented to Wetlands or Other Natural Habitats Reference Notes Amphibians Strong positive response to the proportion of wetlands within 750 to 3000 m of breeding pools Houlahan and Findlay 2003 Studied 74 wetlands in southeastern Ontario Amphibians Distance to natural wetlands was an important factor in predicting amphibian diversity in the restored wetlands, particularly within the first 1000 m Lehtinen and Galatowitsch 2001 Turtles Movement was concentrated within 375 m from the edge of the breeding wetland Roe and Georges 2007 Study based in Australia Green Frog Occurrence increased with the percent forest cover within 1000 m of ponds Mazerolle et al Spotted Salamanders At least 100 m (range: 1.6 to m) around seasonally inundated vernal pools was required for upland migration Veysey et al Wood Frog and Spotted Salamander Positive association with the area of upland forest within 1 km of the pond edge Skidds et al Studied presence of egg masses in ponds Birds Diversity and richness increased with the extent of forest and wetland within 500 m of the wetland edge Mensing et al Birds Presence and abundance of some species linked to the amount of wetland within a 3 km radius Fairbairn and Dinsmore 2001 Study set in prairie marsh habitats 38 How Much Habitat is Enough?

41 Additional considerations, as pointed out by Sheldon et al. (2005), include their observations that amphibians are not randomly distributed in potential habitats in the landscape but tend to be concentrated in suitable habitats that are better connected to each other, and that the presence of terrestrial habitats between wetlands can be an important factor in waterfowl distribution. Semlitsch (2008) also points out that periodic drying of temporary wetlands is an important part of wetland upland forest habitat breeding pool 1 km Figure 9. Wetlands in close proximity with adjacent natural areas their natural cycle that helps reduce amphibian predation by fish and invertebrates, and that alternate or redundant habitats in the landscape can provide critical refuges for amphibians when one breeding pond is disturbed or dries up. While most of the available research on the effects of wetland fragmentation is focused on birds and amphibians, it is intuitive that maintaining hydrologic connections between nearby wetlands (where they exist), as well as wetlands and other nearby natural areas, could also be critical to maintaining their functions. How Much Habitat is Enough? 39

42 2.1.6 Wetland Restoration > Guideline Focus on restoring marshes and swamps. Restore fens under certain conditions. For effective restoration, consider local site conditions, have local sources to propagate new vegetation, and wherever possible refer to historic wetland locations or conditions. Prioritize headwater areas, floodplains and coastal wetlands as restoration locations. > Rationale Defining restoration The terms rehabilitation, creation and restoration apply to distinct activities associated with habitat management and conservation. The term restoration in the general context of How Much Habitat is Enough? encompasses all these activities and refers primarily to the recreation, enhancement or improvement of habitat functions in locations or areas where the habitat had been historically present, but may also capture creation of habitats in alternate locations where conditions are suitable. In the current land use context of southern Ontario, it is simply not possible for many watersheds and jurisdictions to return to estimated historical levels of wetland cover. Also, the ability to restore the diversity and complexity of wetlands, and their wildlife functions, remains questionable, and where possible even partial restoration of a wetland can take many years. Therefore, wetland restoration should only be considered after alternatives for protection have been examined and discarded, as a means to rectify anticipated losses, or when the objective is to increase wetland cover (see Clewell et al. 2004). Regardless, where necessary, targeted and wellplanned wetland restoration can help priority restoration areas meet water quality and wildlife habitat targets at watershed-wide and ecozone scales. Despite the limitations, restored wetlands can provide habitat for a range of species. Lehtinen and Galatowitsch (2001), in their study of amphibian colonization of 12 restored open water wetlands in Minnesota, found that restored sites were colonized by up to 75% of the species occurring in the nearby reference sites within a year. Stevens et al. (2002) found amphibian species diversity was the same between restored and reference sites, and species abundances of amphibians (Northern Leopard Frogs, Green Frogs and Spring Peepers) were higher in restored than in reference sites. Using historical data and reference sites as a starting point for setting realistic targets and identifying appropriate locations for wetland restoration is an ecologically sound approach. Restoration by wetland type Only two wetland types marshes and, to a more limited extent, swamps may be restored with some confidence. 40 How Much Habitat is Enough?

43 Currently, limited information is available on the science of rehabilitating fens and bogs, and the best management strategy (as for all wetlands) is to protect them by protecting their water sources and not altering their watersheds. In some cases, abandoned pits and quarries that are connected to the water table may offer unique opportunities for fen creation (Hough Woodland Naylor Dance Limited and Gore and Storrie Limited 1995). A current review of the state of pit and quarry rehabilitation in Ontario (Skelton Brumwell & Associates Inc. and Savanta Inc. 2009) identifies a number of locations in southern Ontario where rehabilitation of former pits and quarries has involved areas of wetland creation, including a few examples where fen vegetation has become established. Although there is also little published research on successful reproduction of mature forested wetlands, these wetlands are considered less complex (from a restoration perspective) than bogs or fens, and are generally considered reasonable candidates for restoration given the right conditions and sufficient time for trees and tall shrubs to grow. Marshes are considered the most readily restored type of wetland and can be at least partially functional within a few years. As a result, marsh restoration has been widely implemented. Despite this, restoration has often been unsuccessful. Available data for regulated wetland restoration from the United States, from hundreds of projects evaluated over five different states, indicate that less than a third to a quarter were considered successful in terms of area of wetland replaced for area lost. Furthermore, when specific wetland functions were compared between reference and restored wetlands, most restored or created sites had less organic matter, lower plant species diversity and structural complexity, and lower diversity of other groups of wildlife (e.g., amphibians) than their reference sites (Kettlewell et al. 2008; Sheldon et al. 2005). There have even been documented differences in levels of success between different types of marshes, with open water wetlands being most successful, shallow marsh and scrub-shrub wetland restoration being quite successful, and wet meadow restorations being relatively unsuccessful (Sheldon et al. 2005). Common problems included failure of wetland vegetation to establish throughout the site, and lower levels of vegetation and wildlife diversity compared to reference sites. The relatively low level of documented success provides a good rationale for working towards replacement ratios of more than one to one. Another argument for wetland replacement ratio of more than one to one is presented by Gutrich and Hitzhusen (2004), who found time lags for restored wetlands to attain the floristic and soil equivalency of reference wetlands to range from 8 to 50 years. Restoration considerations Successful wetland restoration requires technical expertise. Key variables that need to be considered include soil conditions and fertility (including the presence of organic matter), water level fluctuations, and plant competition and structure (as determined by gradients along wetland edges). The presence of rare species is also important, not just for intrinsic value, but because they indicate the presence of rare habitat conditions that may also be valuable to other species (Keddy 2010; Keddy and Fraser 2002). How Much Habitat is Enough? 41

44 Recommendations for improving the success of wetland restoration include using a watershed approach to target and prioritize site selection, having a good understanding of the local biophysical conditions (especially hydrology, soils, slopes and potential sources for plant propagation), accepting that the restored wetland may not return to its former state or reference condition, maintaining and referring to existing hydrologic (and if possible terrestrial) linkages in the landscape, adding ecological insurance (e.g., try to restore more area, introduce more native diversity and incorporate a greater range of natural gradients), and where possible, undertake large-scale projects where different restoration approaches can be tested (Keddy 2010; Keddy and Fraser 2000; Keddy and Reznicek 1986; Palmer 2008; Sheldon et al. 2005; Verhoeven et al. 2008; Wagner et al. 2008; Zedler 2000). It should also be remembered that humans are not the only mammals playing a role in wetland creation: Muskrats and Beavers influence many wetland functions, and are often active participants in restoration (e.g., Gurnell 1998; Johnston and Naiman 1990; Naiman et al. 1988). A practical process of implementing headwater wetland restoration in agricultural southern Ontario has been developed in Aylmer District of the Ontario Ministry of Natural Resources in southwestern Ontario. The process involves local biologists, drainage superintendents, landowners and others. The methodology incorporates current science, land use considerations, landowner interests and hydrological and biodiversity benefits. A guide book has been produced. This is a valuable strategy for wetland restoration that can be copied across southern Ontario (A. Norman, OMNR, London, pers. comm. 2011). Restoration by location Wetlands restored anywhere within a watershed will provide an array of benefits including regulation of peak water flows and increases in biodiversity, provided that they are in suitable sites. However, the scientific literature is increasingly demonstrating that restoration of wetlands in some areas can be more beneficial than in others. Some guidance for determining the best locations for wetland restoration projects is available (Almendinger 1999; Bedford 1999; DeLaney 1995; Griener and Hershner 1998). Restoration of some major wetland areas such as Great Lakes coastal wetlands can result in extremely valuable ecological benefits; however, these projects can be technically challenging due to their size and complexity. In considering wetland location (see Section 2.1.2), restoration is best targeted to the following areas: Headwater wetlands, particularly swamps, should be restored where they previously existed; On-line or flood plain marshes and swamps should be rehabilitated or restored on second- and third-order streams; Rehabilitation of wetlands in lakes is a high priority because of their extreme importance to fish as well as other wildlife species; and Rehabilitation of wetlands in known historic locations is encouraged, where still feasible. 42 How Much Habitat is Enough?

45 2.2 Riparian and Watershed Habitat Guidelines Lands adjacent to streams and rivers are referred to as riparian. The riparian zone is an area where terrestrial and aquatic systems influence each other (Knutson and Naef 1997), and it functions as an ecotone and ecosystem (Naiman and Decamps 1997). Riparian habitat contains vegetation communities and soils with attributes of both wetland and upland areas, and provides the transition between forest and possibly diminished) by both the aquatic system on one side and the broader terrestrial systems on the other side. Watershed attributes beyond the riparian zone, such as land cover and land use, will also have an influence on stream habitat quality. The habitat guidelines presented here relate to the tributaries of the Great Lakes and St. Lawrence River. The term stream is used here to describe any natural flowing watercourse, though unnatural (i.e., constructed or altered) watercourses may merit similar consideration depending on their function and importance within a watershed. Figure 10. The riparian zone stream, hillside and valley, as well as terrestrial and aquatic ecosystems (Everett et al. 1994, as in Knutson and Naef 1997). The extent of the riparian zone is defined here as the area where vegetation may be influenced by flooding or elevated water tables (Naiman and Decamps 1997), by its related ecological functions, or both. Riparian zones provide two broad types of ecological function. They provide essential services to aquatic habitats as both a buffer between aquatic ecosystems and terrestrial systems, and as contributors of resources including woody structure, nutrients and shade. Riparian zones also provide habitat in their own right, which may be moderated or enhanced (or The focus of this section is principally on terrestrial habitat and its relation to watercourses and wetlands. As such, it does not include in-stream habitat guidance. There is a large and growing body of knowledge on in-stream habitat and hydraulic parameters that should be considered when specifically assessing stream health and considering stream rehabilitation. The width of the riparian zone and the percent vegetated stream length guidelines directly address the amount of riparian area present to provide both direct terrestrial habitat and ecological services to aquatic habitat. It is important to recognize that this entire complex environment requires overarching general protection. This is a complex zone because How Much Habitat is Enough? 43

46 riverine floodplains are species-rich systems containing ecotones at various scales between multiple habitat types (Ward et al. 1999). Fluvial processes in the form of floods and regular variations in water levels also contribute to functional and species diversity within the floodplain (Ward and Tockner 2001). In order to best address stream quality as well as terrestrial and aquatic habitat functions, the area of natural riparian vegetation should encompass the floodplain and upland transition zone or ecotone. Where there is a strong physical disjunct between the stream and upland, such as a bluff, riparian vegetation may have less direct habitat value, although it may have strong value in terms of erosion control and some habitat attributes. Widths necessary to provide effective buffering capability may also be influenced by the sensitivity of the receiving watercourse and its ability to assimilate any stressors. The width and percent vegetation guidelines represent a generic riparian zone that is applicable under the greatest range of geographic, biotic and abiotic conditions. Impervious land cover within the broader watershed will have significant impacts on the quality of aquatic habitat within streams. These impacts may be mitigated to some degree by riparian zones. Relatively narrow riparian zones may be adequate when the broader area is in good condition (i.e., dense, native vegetation on undisturbed soils), and the adjacent land use has low to medium impact potential (i.e., parkland or low density residential). Wider riparian zones may be required to provide sufficient habitat and/or buffering functions for biologically sensitive systems, where the area is in poor condition, where soils are less permeable or highly erodible and slopes are steep, or where the adjacent land use is intense (e.g., intensive row-crop agriculture or urban centres). Finally, measures of water quality and of fish communities provide feedback on the effectiveness of the riparian zone in conjunction with the surrounding watershed land cover in protecting and maintaining the aquatic environment. Fish communities may be affected by direct influences on aquatic habitat such as point source and upstream tributary inputs, or by other in-stream disturbances (human-induced or otherwise), or both. However, the quantity and quality of riparian habitat can help to directly mitigate watershed landscape effects on both water quality and aquatic life. Contributions of the riparian zone to aquatic and terrestrial habitat Stream size and physical characteristics associated with stream order are the products of fundamental biophysical factors (Imhoff et al. 1996; Kilgour and Stanfield 2001; Kilgour and Stanfield 2006; Stanfield and Kilgour 2006). Seelbach et al. (1997) stated that upstream catchment area, geology and slope are the primary determinants of stream size and physical conditions and, allowing for the normal distribution of plants and animals, stream biota. Modifying factors beyond land use such as instream barriers (including dams), channel modifications and point-source discharges will have a significant effect on stream and aquatic community qualities (Stantec 2007). Table 7 shows stream response to human alteration based on underlying conditions. Studies where the effect of adjacent vegetation has been separated from overall catchment vegetation cover show a positive relationship between the overall forest cover and stream health (Wang et al. 2006). Small headwater systems are highly dependent upon vegetative cover for moderation of stream 44 How Much Habitat is Enough?

47 temperature (Broadmeadow and Nisbet 2004) and flow (Swanston 1985) as well as sediment load buffering (Dosskey et al. 2007; Dosskey et al. 2010). Additionally, these systems receive and transport large volumes of beneficial organic matter (e.g., falling leaves and insects), that are processed by fish and benthos downstream (Wipfli 2005). Headwater streams are significantly more efficient at retaining and transforming organic matter than larger streams. The retention and transformation of organic matter upstream affects downstream water quality and the survival and condition of organisms reliant on in-stream food sources (Cappelia and Fraley-McNeal 2007). Drawing from studies on a small shaded stream, Nakano and Murakami (2001) found biomass fluxes between stream and forest accounted for 25.6 and 44.0% of the annual total energy budget of bird and fish assemblages respectively when terrestrial and aquatic ecosystems are intact. In addition, England and Rosemond (2004) suggest that relatively low levels of riparian deforestation along headwater streams can weaken terrestrial-aquatic linkages. In turn, diversity of life in first- and second-order, as well as intermittent, streams contributes to the diversity of life within the entire river and its riparian zone (Meyer et al. 2007). Table 7. General Overview of the Sensitivity of Streams to Human Alteration of Land Cover (based on Stantec 2007) Response Variable Permeable Soil Impermeable Soil Small Catchment Large Catchment Small Catchment Large Catchment Mid-summer water temperature Significant increase Modest increase Modest increase Minor increase Energy supply from inputs Significant decrease Minor decrease Significant decrease Minor decrease Energy supply formed within Significant increase Minor increase Significant increase Minor increase Dissolved oxygen concentration Significant decrease Minor decrease Significant decrease Minor decrease Benthic invertebrates Significant change to more cool and warm water, tolerant forms Minor change to more warm water, tolerant forms Minor change to more warm water, tolerant forms Minor change to more warm water, tolerant forms Fish communities Significant change to more cool and warm water, tolerant forms Minor change to more warm water, tolerant forms Minor change to more warm water, tolerant forms Minor change to more warm water, tolerant forms From a watershed perspective, effective management practices must consider how riparian zones contribute to conditions within local streams especially for stream orders one through three both directly and indirectly, and how they provide terrestrial habitat in their own right. The discussions presented below provide scientifically supported guidance for the How Much Habitat is Enough? 45

48 minimum habitat parameters under which riparian zones can function as aquatic buffers, wildlife corridors and in situ habitats. The definition of the riparian zone within a management context should be flexible enough to encompass these functions as well as address the need for enhanced functions to mitigate potential future impacts. Within watersheds where targets higher than these guidelines can be met and supported, they should be. Watershed land cover and habitat health Watershed land cover beyond the riparian zone does influence stream ecosystems; however, the relationship is difficult to quantify. Allan (2004) and other authors noted that there is an increasing recognition that human actions at a landscape scale affect stream ecosystems (Allan et al. 1997; Strayer et al. 2003; Townsend et al. 2003). This finding supports the use of detailed subwatershed studies in advising stream management decisions. Large areas of forest or other natural cover, and possibly entire catchments, may be required to maintain stream health (England and Rosemond 2004; Harding et al. 1998). However, there has been only partial success at quantifying associations between land use and effects on stream systems given covariation of human and natural influences, mechanisms operating at different scales, nonlinear stream responses, and underlying historical influences (Allan 2004). Riva-Murray et al. (2010) found that while impervious surface explained 56% of the variation in macroinvertebrate communities in the Delaware River basin, secondary land use measures explained an additional 27%. Beyond impervious cover, other potentially important aspects of watershed land cover can include measures such as urban land with tree cover, forest fragmentation (e.g., aggregation index) or aggregation of urban land use. The potential influence of these and/or other similar measures demonstrate the importance of monitoring other aspects of urbanization in addition to impervious surface. Fitzpatrick et al. (2001) found fish communities in good condition with watershed-wide agricultural land cover up to 50% as long as riparian areas contained less than 10% agriculture. Conversely, Stanfield et al. (unpublished) found upland land cover tended not to explain any more variation in fish community composition than could be predicted by riparian buffer composition. Wang et al. (2006) found that fish community composition is more related to watershed land cover above 20% urban and greater than 70% agricultural land cover. In studies in the northeastern Ontario boreal forest and in Wyoming, 25% loss of forest cover led to effects on stream hydraulics and stream substrate composition (Buttle and Metcalfe 2000; Eaglin and Hubert 1993). Allan (2004), in a summary of several studies, noted that there are declines in stream ecosystem health as agricultural land use increases in a watershed. Also, row crops and other more intense uses may have a greater impact on stream health than pasture. Lastly, agricultural landscapes support fewer sensitive fish and insect taxa than forested watersheds. Stream responses in terms of overall ecosystem health vary widely depending upon the study and the nature of the watershed being studied. In the same summary study, urban land use is seen as having a substantial impact on stream ecosystems, more so than agricultural land use. This difference between urban and agricultural land use was similarly observed by Snyder et al. (2003). 46 How Much Habitat is Enough?

49 A number of papers examined links between aquatic ecosystems and percentage of forest cover in the surrounding landscape, and found strong connections between levels of forest cover and aquatic ecosystem health. Johnston and Schmagin (2008) found annual streamflow yields were greatest in Great Lakes basin watersheds with the highest forest cover and topographic relief. Stephenson and Morin (2009) found that forest cover at the catchment scale explained more variation in algal, invertebrate and fish biomass than any other metric. Chang (2006) cited studies that found forested watersheds normally yield streamflow of higher quality than that from other land uses. Other North American studies support the importance of forests for stream health. Cappiella et al. (2005) found healthy aquatic systems in watersheds with at least 45 to 65% forest cover. Goetz et al. (2003) looked at ratings of stream health based on indices of biotic integrity and linked 29.6% tree cover (including trees outside of natural areas) to poor stream health, 37% tree cover to fair stream health, 44.6% tree cover to good stream health and 50.6% tree cover to excellent stream health at the watershed scale. Helms et al. (2009) linked deciduous forest cover of at least 50% with higher macroinvertebrate species richness. Stephenson and Morin (2009) reported that when forest cover was less than 50%, algal biomass was relatively high (but patterns were variable), and that fish biomass began to decline where forest cover was less than 45% at the reach scale. Moreover, higher percentages of porous land cover across the watershed, such as forest, wetland and meadow, will have a positive effect on stream ecology on the basis that they are not impervious surfaces. Forest cover between 40 to 50% has a positive effect on stream ecosystem health, which supports the percentage forest cover guideline in the Forest section of this report (see Section 2.3). Table 8. Summary of Riparian and Watershed Habitat Guidelines Parameter Width of natural vegetation adjacent to stream Percent of stream length naturally vegetated Percent of an urbanizing watershed that is impervious Guideline Both sides of streams should have a minimum 30-metre-wide naturally vegetated riparian area to provide and protect aquatic habitat. The provision of highly functional wildlife habitat may require total vegetated riparian widths greater than 30 metres. 75% of stream length should be naturally vegetated. Urbanizing watersheds should maintain less than 10% impervious land cover in order to preserve the abundance and biodiversity of aquatic species. Significant impairment in stream water quality and quantity is highly likely above 10% impervious land cover and can often begin before this threshold is reached. In urban systems that are already degraded, a second threshold is likely reached at the 25 to 30% level. How Much Habitat is Enough? 47

50 2.2.1 Width of Natural Vegetation Adjacent to Stream > Guideline Both sides of streams should have a minimum 30-metre-wide naturally vegetated riparian area to provide and protect aquatic habitat. The provision of highly functional wildlife habitat may require total vegetated riparian widths greater than 30 metres. > Rationale Vegetation communities within the riparian zone can directly influence aquatic habitat and affect water quality for aquatic life. These functions include moderation of temperature through the provision of shade, filtration of sediments and nutrients, provision of food inputs through organic debris and leaf litter, and contribution to physical habitat in terms of fallen woody material. Vegetated riparian zones also serve as terrestrial habitat and corridors for wildlife as well as places where terrestrial and aquatic food webs interconnect. These diverse functions can be interactive or independent and will vary with watershed context. For example, the ability of riparian vegetation to moderate water temperature may decline with increasing stream width and volume, but it may still provide terrestrial habitat. The riparian zone width requiring maintenance or protection may vary depending on the size (order) of the stream, the steepness of the banks, and the specific management concerns of the local system (USDA 2007). Kennedy et al. (2003) provide detailed tables of riparian widths associated with specific conservation concerns and broader landscape considerations. The 30-metre width guideline provided here is a minimum general approximation intended to capture processes and functions typical of the active riparian zone of a floodplain and the floodplain-to-upland transition with respect to ecological services provided to aquatic habitat. The riparian width guidelines do not directly include transition buffers beyond the riparian zone, but transition buffers should be considered in managing the riparian zone and from an ecosystem management approach. The type of vegetation and other site-specific conditions beyond the immediate riparian zone may be of particular importance in the management of urban watersheds, as urban development entirely changes the characteristic of surface flow that laterally enters the riparian. The effects of vegetation and land cover beyond the riparian zone on stream aquatic habitat are discussed below and in the following section. Also, while adjacent vegetation should be maintained next to lakes for similar reasons as for streams, this guideline was not developed specifically for lakes. In terms of buffering and habitat functions, there are parallels with PZs as discussed in Section 2.1.3, Amount of Natural Vegetation Adjacent to the Wetland, but with a few key differences. Principally, the 30-metre riparian adjacent vegetation guideline is not based on a species- or function-specific need but reflects a general threshold distance for aquatic health and riparian functions. Also, the 30-metre width is meant to capture a variety of protection and habitat functions. And some of the riparian habitat functions, such as wildlife corridors, do not reflect upland habitat needs of aquatic 48 How Much Habitat is Enough?

51 species but the needs of upland species that utilize the riparian system or stream. While the PZ, and to a degree CFZ, concepts can be applied to riparian systems with careful consideration under certain circumstances, it is suggested that, at a minimum, 30 metres of natural vegetation be maintained next to streams. Knutson and Naef (1997) reviewed numerous published sources on varying riparian widths and the effect on stream health. Reported widths ranged from 3 to 200 metres but with prevalence in the 23 to 60 metre range (all of these are applied to both sides of the stream from the edge of the watercourse inland). They concluded by recommending that fish-bearing streams have buffers of either 46 metres or 61 metres (for streams less than or greater than 1.5 metres width respectively), extending to 76 metres for shorelines or streams of state-wide significance. In a later review that reported on studies from across North America and Europe, Broadmeadow and Nisbett (2004) cited ranges of 15 to 70 metres for stream temperature moderation, 15 to 100 metres for sediment removal and control, and 27 to 100 metres for woody debris and leaf litter supply. A partial removal or loss of some riparian trees, however, may not necessarily impair some riparian buffer functions. For example, Wilkerson et al. (2006) found only a 60% canopy cover is required to maintain effective temperature control. In terms of landscape context, Wang et al. (2003) found that land cover within 30 metres of streams in Wisconsin and Minnesota explained more variation in fish assemblages than land covers beyond 30 metres. Frimpong et al. (2005) found that buffers that were 30 metres wide and 600 metres long were the best predictors of the composition of fish communities within Indiana streams. However, watershed land cover, and more specifically the ratio of natural cover (especially, but not necessarily forest cover) to human-dominated impervious land covers, will change the effectiveness of riparian buffers. Roy et al. (2007) found that 30-metre forested buffers protected fish stream communities, but with diminishing returns above 15% urban land cover within the catchment. The effects of catchment land cover are important and are further discussed in subsequent sections. In reviews by Castelle et al. (1994) and O Laughlin and Belt (1995), riparian zone widths of 3 to 200 metres were found to be effective for bank stabilization and sediment erosion control, depending on site-specific conditions. The Castelle et al. (1994) review specifically looked at different riparian widths effects on sediment removal. The relationship between width and sediment removal was non-linear, with disproportionately wider riparian strips required for relatively small improvements in sediment removal. For example, in one test case, widths of 30.5 metres removed 90% of sediments on a 2% slope; however, a width of 60 metres was necessary to remove 95% of sediments on the same slope. On steeper slopes, two other studies found that widths of 60 metres were effective in removing greater than 80% of sediments (Castelle et al. 1994). The frequency and intensity of sediment inputs are important criteria for the effectiveness of the riparian zone in mitigating the effects of sediment inputs. Castelle and Johnson (2000) found that most contributions to aquatic habitat are realized in the first 5 to 30 metres of the vegetated riparian zone (with rooted vegetation contributing to sediment filtering and trees contributing wood structure and shading). In heavy rain events, riparian sediment filtering may be augmented through the use of additional grassy strip areas How Much Habitat is Enough? 49

52 upland of the riparian zone, which can reduce concentrated surface flows (Knight et al. 2010). Knight et al. (2010) observed forest buffers of 17 to 18 metres remained unbreached (i.e., did not develop lateral erosion channels) by concentrated flow events with the addition of 18 to 22 metres of grassy buffer beyond a treed riparian area. Adjacent lands with established vegetation are fairly efficient at removing excess nutrients from surface runoff. In some studies, areas with widths as narrow as 4.6 metres have been 90% effective in removing nitrogen and phosphorus, but most areas require a minimum of 10 to 15 metres. A 30-metre-wide adjacent land area along a stream next to logging operations greatly reduced nutrient levels to better than drinking water standards. Wooded riparian lands in Maryland removed 80% of excess phosphorus and 89% of excess nitrogen, mostly within the first 19 metres. Lee et al. (2003) found that more than 97% of sediment and 80 to 90% of key nutrients could be removed with 16.3 metres of mature grass/woody riparian adjacent land area. Riparian zones as wildlife habitat Riparian systems can provide important wildlife habitat. Habitat may be valuable for its intrinsic values, for example as forested habitat for breeding birds or as habitat for flora or as linear features providing connectivity for terrestrial wildlife movement (Knutson and Naef 1997), rather than any particular relationship to the riparian zone itself. Corridor and habitat widths for mammals, reptiles and amphibians are often dependent on the requirements of individual species, and these are discussed elsewhere in this document. However, it is worth noting that widths to address ecological concerns are much wider than those recommended for water quality concerns (Fischer 2000; Fischer and Fischenich 2000). As the width increases, factors such as overall habitat heterogeneity become important and the habitat requirements of species exceed the area of the riparian zone. Riparian areas, specifically due to their association with water, provide core habitat areas for many herpetile (reptile and amphibian) species. Semlitsch and Brodie (2003) suggest that the 15 to 30 metres of adjacent land often prescribed as a buffer to protect wetland species is inadequate for amphibians and reptiles, as riparian areas are not buffers per se but rather are core terrestrial habitats used by these species. The minimum suggested riparian zone requirements for some herpetiles ranged from 127 to 205 metres. There is a wide range of suggested appropriate riparian widths based on function, with the published range in the literature varying from a few metres to over 100 metres, depending on the study and level of representation and confidence (e.g., 95% occurrence within 175 metres). The recommended 30-metre width is supported in the literature as a general guideline minimum for many riparian systems. The 30-metre guideline may provide for basic terrestrial habitat; however, a greater width may be required to provide for a highly functional wildlife habitat. 50 How Much Habitat is Enough?

53 2.2.2 Percent of Stream Length Naturally Vegetated > Guideline Seventy-five percent (75%) of stream length should be naturally vegetated. > Rationale This guideline focuses on the cumulative effect of the riparian zone on stream habitat, and water quality as related to stream habitat. As described, riparian zones contribute to stream habitats in many ways. At the local scale, natural but otherwise open landscapes (i.e., not forested) require a forested riparian zone of at least 150 metres in order to generate substrate habitat capable of supporting benthic communities other than those tolerant to human disturbance (Wooster and DeBano 2006). A Toronto-area study found stream degradation occurred when riparian vegetation amounted to less than 75% cover along the length of firstto third-order streams (Steedman 1987). Alternatively, in a field test of this guideline, the Toronto and Region Conservation Authority commented that there are many cold water streams that have less than 75%, or even less than 50%, vegetated riparian habitats, pointing to an ability for some streams to resist degradation below the 75% threshold. Related comments were provided by Gartner Lee Limited (1997a) in a field test in Hogg Creek located in the Severn Sound Area of Concern, Ontario. In Hogg Creek, only 43% of the firstto third-order streams had vegetated riparian zones. Several tributaries of the main branch of Hogg Creek exhibited cold water characteristics that seemed to relate to a high ratio of baseflow (approximately 47%) as a percentage of average annual discharge per square kilometre. Gartner Lee Limited (1997b) also noted that the presence of cold water streams is heavily dependent on the geological characteristics of the area. Riparian vegetation provides proportionately greater benefits to stream aquatic habitat along the headwaters of streams. There is growing literature on the interaction between stream and adjacent terrestrial communities (Iwata et al. 2010; Nakano and Murakami 2001). From a watershed perspective, planting vegetation along smaller systems (i.e., less than third-order) will produce greater aquatic habitat benefits than planting along higher-order rivers, although both provide benefits. Vegetation along a smaller stream may have a greater potential to provide sufficient cover in order to lower the summer maximum stream temperatures than along the banks of a large river, but more importantly, aquatic/terrestrial biomass exchange may be more balanced in smaller streams with abundant adjacent cover. Given that the form and function of headwater streams are strongly influenced by the character of adjacent lands, a lack of adjacent natural vegetation can result in dramatic alterations in flow and sediment regimes. How Much Habitat is Enough? 51

54 Riparian vegetation, however, will still be of great value along larger rivers for species such as waterfowl, reptiles and amphibians, as well as mammals such as North American River Otter, American Mink and Beaver. In this context, there is a similarity to wetland CFZs (see Section 2.1.3): within riparian zones, there will be species-specific areas of adjacent land required to complete life cycles for species that spend their life cycle in both aquatic and terrestrial habitats. Lower-order streams comprise the majority of stream systems in the lower Great Lakes basin. For example, approximately 75% of the length of the Credit River system is comprised of first- to third-order streams, 83.5% of Duffins Creek is first- to third-order streams, and 75% of Carruthers Creek is first- to thirdorder streams (TRCA 2012; CVC 2012 pers. comm.). 1 st order 1 st order 2 nd order 2 nd order 3 rd order 4 th order The 75% guideline is presented as a minimum, given the literature and the fact that headwaters in southern Ontario tend to comprise approximately 75% of stream lengths. Effective management approaches should consider local watershed conditions (land cover) when applying this guideline, as discussed in the following section. 3 rd order Figure 11. Stream order, showing extent of first- to third-order streams 52 How Much Habitat is Enough?